Talk:Raspberry Pi: Difference between revisions

Content deleted Content added

Inline

Revision as of 14:16, 6 March 2012

This is the talk page for discussing improvements to the Raspberry Pi article.
This is not a forum for general discussion of the article's subject.

Put new text under old text. Click here to start a new topic.
New to Wikipedia? Welcome! Learn to edit; get help.

Article policies

Find sources: Google (books · news · scholar · free images · WP refs) · FENS · JSTOR · TWL

This article is written in American English, which has its own spelling conventions (color, defense, traveled) and some terms that are used in it may be different or absent from other varieties of English. According to the relevant style guide, this should not be changed without broad consensus.

This page is not a forum for general discussion about Raspberry Pi. Any such comments may be removed or refactored. Please limit discussion to improvement of this article. You may wish to ask factual questions about Raspberry Pi at the Reference desk.

Computing C‑class Mid‑importance

	This article is within the scope of WikiProject Computing, a collaborative effort to improve the coverage of computers, computing, and information technology on Wikipedia. If you would like to participate, please visit the project page, where you can join the discussion and see a list of open tasks.ComputingWikipedia:WikiProject ComputingTemplate:WikiProject ComputingComputing articles
C	This article has been rated as C-class on Wikipedia's content assessment scale.
Mid	This article has been rated as Mid-importance on the project's importance scale.

Archives

Archive 1	Archive 2	Archive 3
Archive 4	Archive 5	Archive 6

This page has archives. Sections older than 31 days may be automatically archived by when more than 4 sections are present.

Shortcomings?

Lol, I have been reading other wiki articles of Raspberry Pi copies/clones such as the BeagleBoard etc and none of them have a 'shortcomings' section and don't believe that the Raspberry Pi needs one either. Twobells (talk) 17:11, 1 December 2011 (UTC)[reply]

The board lacks a battery-backed real-time clock (RTC), and documentation for the Broadcom SoC isn't available to the public. (removed text)

lolz back at ya. Just because other articles don't have such a section, doesn't mean that such a section shouldn't be included. • Sbmeirow • Talk • 18:19, 1 December 2011 (UTC)[reply]

The information is allowable. There is no requirement to do what other articles have done.

That being said, you can't call it "shortcomings" or list what you think are shortcomings. Please see Wikipedia:No original research for the reasons why. You need to call the section "Criticism" and you need to provide citations to reliable sources for each criticism. See Wikipedia:Citing sources and Wikipedia:Identifying reliable sources. --Guy Macon (talk) 18:41, 1 December 2011 (UTC)[reply]

I've seen positive and negative criticism in "Reception" sections, so that might be the best place to put it. The board lacking a RTC has been stated in some articles already, plus it is obviously missing, per the table. • Sbmeirow • Talk • 18:59, 1 December 2011 (UTC)[reply]

Summary Of Limitations

Real-time clock (RTC) isn't battery-backed, thus it doesn't know the current time or date when powered up.
ARM11-core doesn't have an integer hardware divide instruction. http://wanderingcoder.net/2010/07/19/ought-arm/
Broadcom BCM2835 IC datasheet isn't available (requires NDA). http://www.broadcom.com/support/
The GPIO connector is not keyed, nor marked, to prevent newbies from plugging in a cable in the opposite direction.
Mounting hole issues.

• Sbmeirow • Talk • 21:20, 21 December 2011 (UTC)[reply]

A reception section would have to neutral and be based on reliable sources reviewing the product, but so far none have been shipped to be reviewed. Shortly after the first boards ship would be an ideal time to research and write such a section. --Guy Macon (talk) 02:10, 29 December 2011 (UTC)[reply]

Amazing, hell must have froze over, because Broadcom has released a partial datasheet to Raspberry Pi group.

Read about it http://www.raspberrypi.org/archives/615

Here it is http://dmkenr5gtnd8f.cloudfront.net/wp-content/uploads/2012/02/BCM2835-ARM-Peripherals.pdf

• Sbmeirow • Talk • 00:57, 7 February 2012 (UTC)[reply]

New Photo Request

The current photo is the old alpha board, thus someone needs to upload photos of the beta or production Model A and Model B boards. • Sbmeirow • Talk • 14:50, 3 January 2012 (UTC)[reply]

Know of any free photos of the production boards? -- Trevj (talk) 19:50, 3 January 2012 (UTC)[reply]

I can't believe that nobody at the Raspberry Pi foundation reads this page. If you are with the foundation, is everyone there aware that you have chosen to restrict Wikipedia's use of the pictures and text on your website because you are not using the Creative Commons Attribution-ShareAlike License? If you meant to do that, that is of course your right, but I can't help wondering if perhaps you simply are not aware of the consequences of your licensing choices. --Guy Macon (talk) 20:46, 3 January 2012 (UTC)[reply]

I added a Beta board picture, I hope there won't be any licensing problems, I'm a bit new to this... Ideeman1994 (talk) 22:46, 3 January 2012 (UTC)[reply]

EDIT: I just asked Liz, and she said we could do it without any problem... Check here, search for "Ideeman1994": http://www.raspberrypi.org/archives/498#comments Ideeman1994 (talk) 23:12, 3 January 2012 (UTC)[reply]

I am pretty familiar with how copyright on Wikipedia/Wikimedia works and I have the file on my watchlist and will take care of any problems that crop up.

It's actually technically wrong at the moment (the link at raspberrypi.org shows that permission has been given, not that it is in the public domain) but that's good enough for now given that the same comment says it will be CC BY-SA 3.0 licensed soon.

I would like to emphasize that the best way to handle this is to put as much of the content at raspberrypi.org as possible under the Creative Commons Attribution-ShareAlike license (CC BY-SA 3.0). A good way to do this is to put this statement somewhere on the website: "unless otherwise specified, all content on this website is licensed under the Creative Commons Attribution-ShareAlike license", then mark anything you want to keep proprietary and closed as "Copyright 2012 Raspberry Pi Foundation". Please pass this info on; I haven't had a lot of luck getting a response from the foundation. --Guy Macon (talk) 23:43, 3 January 2012 (UTC)[reply]

Someone should update the license fields for the photo. Uploading photos is kind of a pain, especially if you want to be 100% sure that someone else won't delete the photo because they don't think you have the proper copyrights. I wrote up my procedure of what I do...and I've had ZERO problems...likely because I upload to Flikr and set rights to my photos over there first. See User:Sbmeirow/HowToUploadPhotosToWikipedia for tips and ideas. Note this is a user article. • Sbmeirow • Talk • 00:20, 4 January 2012 (UTC)[reply]

Don't worry. It won't be deleted while I am watching it -- I will intervene if deletion is proposed. Let it be for at least a week. If during that time Liz releases it under CC BY-SA, (or, better yet, follows my advice above) whover notyices it, please post that info here and I will do the rest. --Guy Macon (talk) 00:30, 4 January 2012 (UTC)[reply]

When the various photos have the appropriate licenses, IMO previous versions of the board could be depicted in the History section. -- Trevj (talk) 08:16, 4 January 2012 (UTC)[reply]

I hope I didn't do anything wrong... I was a bit lost, didn't know what license I had to use, considering what Liz said :) Ideeman1994 (talk) 10:04, 4 January 2012 (UTC)[reply]

You done did good. <smile> The great thing about Wikipedia is that we are all encouraged to be bold and make improvements even if we aren't quite sure what the proper procedures are, leaving a note so that someone else can review and improve it. --Guy Macon (talk) 17:47, 4 January 2012 (UTC)[reply]

I have changed the infobox picture to one, from wikicommons, showing a beta board in use, that is freely available, and used in other wikipedia, such as French, German and Russion versions. That should resolve any free access issues. Regards, Lynbarn (talk) 16:32, 7 February 2012 (UTC)[reply]

How about adding a coin in the image for scale? — Preceding unsigned comment added by 122.167.214.38 (talk) 17:01, 4 March 2012 (UTC)[reply]

I would suggest a British coin and a US coin, with the British coin partially covering the US coin. This would be a subtle nod to the fact that this is a UK product that is meant to be used world-wide.^{Struck out comment- I didn't know it was against policy} --Guy Macon (talk) 19:11, 4 March 2012 (UTC)[reply]

err -it's the size of a credit card - I imagine that more people in the world are familiar with the size of a ceredit card than with either countries' coinage! :) Regards, Lynbarn (talk) 20:23, 4 March 2012 (UTC)[reply]

Wikipedia has a policy specifically -against- using a coin to depict the size of an object, precisely because of such issues. Besides for an initiated person (a typical first buyer) of an RPi there are enough visual clues to the real size of the device, for example the Model-B USB connector, , the 2.54mm pinheader, and the RCA connector provide clear optical references. Mahjongg (talk) 00:26, 5 March 2012 (UTC)[reply]

Interesting! I agree that coins aren't the best way to communicate relative size to an international audience, but where can one read that policy? --Ds13 (talk) 01:25, 5 March 2012 (UTC)[reply]

Took me some time to find it again, but look at this:

Mahjongg (talk) 11:31, 5 March 2012 (UTC)[reply]

That's useful. Although the preliminary beta photos used a coin, the above makes perfect sense. I think use of a ruler (with both imperial and metric measurements, as in the example) would be best. -- Trevj (talk) 12:45, 5 March 2012 (UTC)[reply]

Hey, that's cool. Thanks for finding the template. Good to keep in the editor's palette for future. --Ds13 (talk) 03:53, 6 March 2012 (UTC)[reply]

Risc OS

The article RISC OS is stating that Raspberry Pi supports or supported Risc OS, but in this article it states that it's unsupported. I didn't read the articles fully but it looks like contradictory information. — Preceding unsigned comment added by 194.180.57.12 (talk) 11:39, 10 January 2012 (UTC)[reply]

I agree that it's contradictory. The ambiguity isn't helped by the OSNews article being titled "Raspberry Pi To Embrace RISC OS". This isn't strictly the case. They provided a loan board for the show in London and coincidentally a Broadcom employee happens to be a RISC OS developer (this latter information has been mentioned on the RPi forum). -- Trevj (talk) 13:57, 10 January 2012 (UTC)[reply]

The Raspberry Pi FAQ is used as a citation to state RISC OS as supported platform, but there's nothing definitive in the FAQ about it, just a link to a BBC article claiming the same thing whose author could have just lifted the claim from this article. I suggest removing that citation. I also wouldn't rely very strongly on OSNews as a source, either. You really need to have someone actually associated with the project stating fact, rather than some gossip site claiming something, potentially based (again) on what they read here. PaulBoddie (talk) 00:10, 18 January 2012 (UTC)[reply]

I've made a couple of edits which should help to address this. The BBC News article makes no reference to RISC OS, apart from in the spec box. There appears to be offical acknowledgement (JamesH is a forum admin) of the RISC OS work, but that post isn't ideal as a reliable source, and it's a primary source anyway. -- Trevj (talk) 10:56, 19 January 2012 (UTC)[reply]

I have removed RISC OS from the infobox. The infobox and "specification" sections should agree, but they didn't. The "specification" section omitted RISC OS and footnoted some background. I don't care either way, but it's confusing for a new reader to see inconsistency in these lists. --Ds13 (talk) 23:24, 23 January 2012 (UTC)[reply]

I agree that consistency is desirable for readers. I'm not going to revert the infobox and spec table myself now. This edit could have used a more explicit edit summary, but the extra note is a valid addition. -- Trevj (talk) 09:18, 24 January 2012 (UTC)[reply]

This article on zdnet.de claims Der Nutzer kann zwischen Linux und RISC OS als Betriebssystem wählen (Google translate to English). It doesn't seem to say how the user chooses, so this has probably been inferred by the authors from the expected availability of the port. The choice may well be along the lines of "Use this alternative SD card instead." -- Trevj (talk) 12:17, 9 February 2012 (UTC)[reply]

Copyright vs. Creative Commons licensing

File:Raspberry Pi Logo.svg is still a proprietary / copyrighted image, and much of the material associated with the Raspberry Pi project is also proprietary / copyrighted. I am seriously considering adding a section to the article specifying what is and isn't free. This would be a lot easier if the foundation specified that unless otherwise specified everything is under a CC-BY-SA 3.0 License, but my requests to the Raspberry Pi Foundation asking them to do that were never answered. Does anyone have contact info so I can make one last plea before opening this particular can of worms? --Guy Macon (talk) 19:14, 30 January 2012 (UTC)[reply]

We have a problem that will make it difficult to bring this article up to Wikipedia's standards. The problem is evidenced by the above sitting for 9 days with no response. --Guy Macon (talk) 11:14, 8 February 2012 (UTC)[reply]

Template:third-party

Following a review of the article, which followed a reversed edit of a paragraph I deleted, I believe the sources cited throughout are very close to the subject, and thus do not give a balanced or neutral POV. Please comment herein. Regards. Lynbarn (talk) 00:22, 7 February 2012 (UTC)[reply]

I have asked any wikipedians on the official Raspberry Pi forum to lend a hand at generally improving the article. Regards Lynbarn (talk) 11:23, 7 February 2012 (UTC)[reply]

I think this tagging is well-intentioned and agree that there are a few references to primary sources. However, a different tagging approach would be more helpful IMO. For example, {{Primary sources}} is also available. Some distinctions between that and {{Third-party}} were discussed at Wikipedia:Templates for discussion/Log/2011 August 4#Template:Third-party. Or are you concerned that the many secondary sources used are in some way linked to the Foundation and are therefore not independent?

I think it would be better to use specific tagging of identified sources with {{Self-published inline}} or {{Third-party-inline}} as appropriate. Then editors can seek out secondary sources as replacements, if available. -- Trevj (talk) 09:09, 8 February 2012 (UTC)[reply]

I haven't done much in terms of tagging articles, and {{Third-party}} seemed the closest I could find to what I felt was wrong. My concerns are that many refs. are to comments in fora and blogs closely associated with the official website, but not necessarily endorsed by the Foundation, and also there are comments that don't accurately reflect their cited sources. I will certainly look at the tags you've mentioned - Thanks for pointing them out! Regards, Lynbarn (talk) 11:06, 8 February 2012 (UTC)[reply]

A related problem is that there simply are not a lot of reliable third party sources for products that have no yet started shipping. I expect a lot more coverage once the boards are available for sale. --Guy Macon (talk) 11:08, 8 February 2012 (UTC)[reply]

I agree that a comment from the forum isn't reliable unless it's known that the poster is directly connected with the project (but basing aspects of the article on such comments may be original research). The blog would IMO qualify as a reliable source, although it's a primary source and so may only be used on Wikipedia to make straightforward, descriptive statements of facts that any educated person, with access to the source but without specialist knowledge, will be able to verify are supported by the source. But as Guy Macon points out there's not (currently) a lot of coverage in the press for some aspects. -- Trevj (talk) 12:19, 8 February 2012 (UTC)[reply]

The time has come to put the Crystal Ball into storage

Wikipedia is not a crystal ball. We need to stop putting claims like "Scheduled for introduction in November 2011" (now "updated" to "Scheduled for public release in January 2012", and due for the next "update" in a couple of days). WP:SPECULATION is quite clear: "Individual scheduled or expected future events should only be included if the event is ... almost certain to take place." It is time to stop playing Whac-A-Mole with these predictions and to start reporting things only after they actually occur, like we did with the eBay auction. Other speculation that should be removed include statements such as "shipping versions are planned to be credit card sized." If a prototype was built and it was the size of a credit card, report that, but don't speculate on the size of what will eventually ship. Likewise, "KOffice and Python are bundled with the Raspberry Pi" is something that may change, so we should not include speculation about it.

There are some things that are almost certain to be true about the boards when they ship and thus can be included in the article. One example: "It does not have a built-in hard disk drive, instead relying on an SD card for long-term storage." There is a vanishingly small chance that between now and when the boards ship the design will change to use a hard disk instead of a SD card, so it's OK to say that "It does not have a built-in hard disk drive."

Let's take out the speculation and report only verifiable events that have already happened or are almost certain to happen, and let's report only features that are known to be on prototypes, reporting them as "the prototype has..." if they may change. --Guy Macon (talk) 18:56, 30 January 2012 (UTC)[reply]

Articles are not limited to events that have already happened. As the WP:CRYSTAL policy above states, "Predictions, speculation, forecasts and theories stated by reliable, expert sources or recognized entities in a field may be included, though editors should be aware of creating undue bias to any specific point-of-view. In forward-looking articles about unreleased products, such as films and games, take special care to avoid advertising and unverified claims." So WP editors themselves should not be predicting nor speculating, but as long as this article has an NPOV, then it's just fine to cite speculation and predictions from reliable sources. --Ds13 (talk) 07:22, 3 February 2012 (UTC)[reply]

The Rasberry Pi foundation is not a reliable source on the subject of when the boards will ship or what software will ship with them. This is evidenced by the various past predictions that were later revised. This is in no way a criticism of the RPF - most entities involved in electronics manufacturing are unable to accurately predict whether they will encounter delays --Guy Macon (talk) 10:56, 8 February 2012 (UTC)[reply]

...and to be fair, the foundation has never suggested a firm delivery date - they have expressed hopes that they would be able to deliver within certain timeframes, but these have been mis-interpreted and solidified by observers. In truth, there IS NO reliable source on the subject, so it may be better to avoid repeating the mistake herein, by leaving the delivery date open. Regards, Lynbarn (talk) 11:13, 8 February 2012 (UTC)[reply]

Is there broad agreement that the product release will almost certainly *happen* some time? I've not seen anyone saying "this may not make it to market". --Ds13 (talk) 17:51, 8 February 2012 (UTC)[reply]

I think there is, but as nobody - even the manufacturer - knows when, there is no point including a non-statement in the article.

I wouldn't say "almost certainly." Organizations closing down are not unknown, and there are external factors that are likely but not "almost certain." For example, Broadcom might suddenly stop making the BCM2835 SoC. Unlikely, but not impossible. Fortunately, we have a solution for this sort of issue, which is already in the article. Just say "The foundation plans to release" rather than "The foundation will release." Although the Rasberry Pi foundation is not a reliable source on the subject of when the boards will ship or what software will ship with them, they are a reliable source for what they plan to do. --Guy Macon (talk) 03:10, 9 February 2012 (UTC)[reply]

Yup, agree. And I'm always in favour of qualifying *everything*, such as "proposed USB port", "proposed price", "scheduled date", etc. It's heavy, but this lets verifiable plans get through while still emphasizing that *anything* can still go sideways before release. --Ds13 (talk) 07:26, 9 February 2012 (UTC)[reply]

deleted material and protected article

User Lynbarn has twice and user Macon once deleted notable and well sourced material that I wrote and that user Sbmeirow restored the first time. Tom Morris in protecting the page refers to "unsourced/badly sourced" material, but every sentence has a precise citation to 4 independent sources, all of whom corroborate the others. Lynbarn claims that the material may be unreliable because one of the 4 sources uses a pseudonym on the R-Pi forum, and that R-Pi forum members are not neutral third parties. But this is the pot calling the kettle black. Lynbarn is an unabashed cheerleader for the R-Pi foundation. His user page says: "I discovered a fantastic British computing initiative with massive potential - the Raspberry Pi Foundation." He himself is one of the R-Pi forum's top posters, with 159 posts as of today. He acts as a self-appointed R-Pi forum administrator, dispatching suspected trolls. ^[1] User Lynbarn apparently feels it is his role to keep the wikipedia article the way the R-Pi publicist would want it. He raised the issue on the R-Pi forum, and the R-Pi publicist Liz, wife of the R-Pi ceo, responded: "Very pleased to see that Lynbarn has taken out some of the references to this forum, which were being used as source material; I'm a Wikipedia editor myself, and I'd have stomped all over it if it was on an entry I was allowed to edit!"^[2]It seems Lynbarn is doing the dirty work that Liz is unable to do herself as a representive of the R-Pi Foundation. Liz and Lynbarn use the excuse that references to the forum should not be used as source material, on the presumption that forum poster would be too close to the R-Pi, and therefore biased in its favor. And in fact two of the sources do work for Broadcom, the R-Pi's close partner and SoC supplier, and the other two are enthusiastic supporters. But their words were being used not as puffery or cheerleading for the R-Pi, but to temper and balance the R-Pi foundation's fundamental claim that the device is intended for, and is suitable for, developing software, particularly Python software, on the device itself. In a recent published interview, R-Pi trustee David Braben is quoted as saying: "The Raspberry Pi Foundation hopes to offer children (of all ages), the opportunity to learn hands-on computing skills by creating software on the low cost, credit-card sized device." Perhaps the Broadcom empoyee was exagerating when he said it took 5 hours on a R-Pi to compile an app that took only 10 minutes to compile on his PC. The app's own users's guide says it should only take 5 minutes to build on a "P4 1.7Ghz/512 Mb." But it is unlikely he would be exaggerating in the opposite direction, trying to make the R-Pi look worse than it actually is. So Lynbarn should be happy about any possible insider bias in this case, not afraid of it.— Preceding unsigned comment added by 72.211.206.238 (talk) 17:53, 11 February 2012 (UTC)[reply]

I am going to assume that you are referring to this edit. It would help if you were to provide links to diffs as i did above. If you don't know how to do that, just post the exact time and date of the edit you are referring to and I will insert the link for you.

Re: "User Lynbarn has 3 times so far deleted notable and well sourced material that I wrote and that user Sbmeirow restored the first time", it does not appear that the material is the same. See link 1 and Link 2.

I will address the meat of your argument later; the above is just some housekeeping so that we are all talking about the same thing. --Guy Macon (talk) 22:18, 11 February 2012 (UTC)[reply]

You are correct that there were two versions of the deleted material (and your links are correct). The original was first removed on 11:06 7 February 2012.[4] The second (improved) version was removed on 11:58 9 February 2012.[5] 72.211.206.238 (talk) 22:40, 11 February 2012 (UTC)[reply]

Just my two cents, but 224 MB (256BM - 32MB for a video framebuffer) should be plenty for a rudimental OS (based on the Linux kernel and X-windows) and a (Python) compiler. The reason its not running (fast) now is that they are probably running it on a complete unmodified Linux distro with all the overhead, using so much memory that the system is forced to use virtual memory, and virtual memory on a SD-card is bound to be extremely slow, even when the system supports faster (non SPI) I/O with the SD-card, swapping will take ages. Recent Linux distro's are simply not "lean and clean" enough for these kind of systems anymore, except perhaps for special implementations like puppy linux that do not use a large and memory hungy desktop manager like GNOME. I predict that a reasonable python implementation will run fine on the Raspberry, if they simply only implement the software that is really necessary, leaving out all the unnecessary cruft. Modern programmers are simply spoilt with resources. It can't hurt to learn them from the beginning to be a bit more frugal. Mahjongg (talk) 01:27, 12 February 2012 (UTC)[reply]

Hi Mahjongg. I am reminded of this guy who can get 59 miles per gallon in a honda accord, but he has to jetison everything not bolted down, tailgate behind 18-wheelers, and wishes he could lose 60 pounds.^[3] Yes, you can do it, but you wouldn't want to do it on a daily basis, and you wouldn't advertise an accord as getting 59 mpg. More seriously, if you reduce the GPU to 32MB, can you still get accelerated 2D graphics, like X-windows? I presume someone doing programming would also occasionally want to use a pdf viewer and/or web browser to look up documentation, so you'd want those to be accelerated. I don't think you can adjust the GPU memory allocation without rebooting, and you don't want to do that very often because it takes forever. Some fun and educational things to do in Python are not even remotely possible, like the "natural language toolkit", nltk, with multi-gigabyte corpora. But even with more modest programs, Python can use up memory in a hurry. — Preceding unsigned comment added by 72.211.206.238 (talk) 04:16, 12 February 2012 (UTC)[reply]

I am on a hot project right at the moment and don't have the time to give it my full attention, but I do intend to analyze this and weigh in on the content dispute. Right now I am supporting neither side. In particular I want to examine the sourcing before I express an opinion. If everybody involved would re-read WP:RS and WP:V, that would be a big help.

I do have time to chip around the edges a bit while waiting for a test run to be set up, so let me comment on some side issues.

First, could we all please indent properly and sign our comments? Just look at what you are replying to, count the colons(":") at the start, and use one more colon. Then end the comment with four tildes ("~"). Thanks!

Second, don't make assumptions about how people will use the Raspberry Pi. For example, I am very likely to be using it headless - no monitor or keyboard, running a single embedded program. I have several uses where an Arduino is not quite powerful enough and I want to replace a 100Mhz 486DX system with 64MB of RAM with something modern.

Third, be careful about speculations on what the RP is likely to be able to do. Read the section above about crystal balls again. In particular, I would like to respond to the above comment "...to temper and balance the R-Pi foundation's fundamental claim that the device is intended for, and is suitable for, developing software, particularly Python software, on the device itself." Wrong, wrong wrong! This is a completely wrong kind of thinking. We are not to "balance" the claims of the RPF. If they claim it cures cancer, we simply report that they claimed that. If some other reliable source claims that the cancer claim is wrong, we report that as well. We don't say whether or not the claim is true.

Fourth (and I haven't dug into the details, I am just responding to a couple of things that caught my eye in the comments above) blogs and online discussions are not as a rule considered to be reliable sources. Again please study WP:RS. --Guy Macon (talk) 06:42, 12 February 2012 (UTC)[reply]

I noticed two mistakes in what I wrote here. I claimed that user Lynbarn deleted the material 3 times, but the second deletion, was instead by user Macon, on 10:42 8 February 2012. I also said that Lynbarn claims the material is unsourced. That should say that Tom Morris in protecting the page refers to "unsourced/badly sourced". My appologies to Lynbarn for the mistakes. I have taken the liberty of correcting the original, in order to minimize any further damage or confusion. 72.211.206.238 (talk) 08:39, 12 February 2012 (UTC)[reply]

Lynbarn's response

Yes, I did twice revert a section of text from this article, and I gave what I consider to be two reasonable and reasoned explanations for the reversals. I did not break the WP:3RR rule, so I was (As far as I am concerned) perfectly justified in making those reversions. I have also made other changes, in an attempt to improve the standard of the article (in Wikipedia terms, using the MoS and other wiki resources as my guide) to make it into a more general enclopedia article. I still believe that the article relies too much on a small number of sources, including the Raspberry Pi Foundation's own forum (more of that in a moment). It was not and is not my intention to make the article biased for or against the project, but to make it wikipedia standard-compliant, neutral, informative and generally of interest. Wikipedia is not the personal fiefdom of the editors, and I welcome all those who like me, wish to improve the quality of this and any other article.

It was also I who requested that the article be semi-protected, and I gave a reason. A wikipedia Admin reviewed my request, and acted accordingly, In truth, I was surprised he protected the page for a full month, as I had expected perhaps a week, to protect the article and give an opportunity for myself and others to continue the tidying exercise.

There are a number of comments in the text above which relate personally to me, and I will therefore address them accordingly:

Lynbarn is an unabashed cheerleader for the R-Pi foundation.

Nothing is perfect and where I find fault, I will criticise in a constructive and civil manner - it is my way. - although I have no link to the Foundation, and no inside information, I do happen to believe it is generally a GOOD THING.

His user page says: "I discovered a fantastic British computing initiative with massive potential - the Raspberry Pi Foundation."

Yes, I did, and I still think its a fantastic initiative - why is that relevent to this discussion?

He himself is one of the R-Pi forum's top posters.

Yes, but again, why is that relevent? Not all my posts are pro the Foundation, or the project.

He acts as a self-appointed R-Pi forum administrator, dispatching suspected trolls.

No, I don't, I act as a responsible member of an on-line community, making my views known, answering questions when I know the answers, and being civil and friendly to all I encounter, just as I do on Wikipedia. What I actually said here was:

Maybe you got off on the wrong foot, but 10,000 others seem to be quite happy with the way the Admins perform their often thankless tasks.

Just take a step back, breath deeply, count to ten, then start again. Do bear in mind though, that not all the Admins are involved with the Foundation in any other way than as Admins to this forum.

Far from "dispatching trolls", this was merely some advice to a new member of the forum, who was having some difficulties getting his point across to other members. It would also help to read it in the context of the other posts on that thread, but here is not the place.

User Lynbarn apparently feels it is his role to keep the wikipedia article the way the R-Pi publicist would want it. He raised the issue on the R-Pi forum.

Yes, I did raise it there, for precisely the same reasons I edited here - because the article was not of good quality (in Wikipedia terms). I said:

The Raspberry Pi has an article on Wikipedia. For many people, this is likely to be their first place of reference when they hear about the project.

To my mind though, as it stands currently, the article is in need of some reworking to bring it up to an acceptable wikipedia standard. At present, it relies heavily on references to this forum, the blog and other sources very close to the project, and as such contravenes several of the Wikipedia guidelines. Are you Wikipedia editor? Could you help to develop the article in line with the general wikipedia principles? I have already been involved in a difference of opinion with one wiki editor over one particular paragraph, and it would help to get consensus on the best way to develop the complete article into a trustworthy, objective, complete and well-written article.

Again, this is not really the place, but I feel it necessary to respond to the comments made about me above.

Note, there is no mention of favouring the project, merely of improving the article. Liz did indeed respond afterwards - I had not known until then that she too was a Wikipedian.

Every edit page contains the following comment to editors:

If you do not want your writing to be edited, used, and redistributed at will, then do not submit it here. All text that you did not write yourself, except brief excerpts, must be available under terms consistent with Wikipedia's Terms of Use before you submit it.

All editors would be well served by reading and remembering this basic tenet of Wikipedia. Regards, Lynbarn (talk) 19:12, 13 February 2012 (UTC)[reply]

Oh - and I should add, I have today been asked to act as Admin on the aforementioned forum, so although I will continue to monitor, and perhaps make quality improvements to this article, Wikipedia POV guidelines indicate that the majority of improvements to this article from now on may need to be applied by fellow wikipedians. I do hope, trust and believe that the result will be an article that is trustworthy, objective, complete and well-written. Lynbarn (talk)

Lynbarn has addressed my concern that he does not have a neutral POV, confirming that he is now an official administrator of the R-Pi forum. He has also demonstrated his comfort in using the Argumentum_ad_populum fallacy against anyone who dares to disagree with the 10000 R-Pi forum members. However, he has not addressed the main issue of why the non-self-serving first-hand published reports of his fellow R-Pi forum administrators, including Broadcom employees JamesH and Dom, should be considered unreliable third-party reports, unsuitable for Wikipedia. 72.211.206.238 (talk) 16:11, 15 February 2012 (UTC)[reply]

How about discussing content, not throwing mud at editors? It's also a bit rich to attack one editor who has been clear about their involvement with the Raspberry Pi, whilst you remain an anonymous spa. Andy Dingley (talk) 16:28, 15 February 2012 (UTC)[reply]

72.211.206.238,"Argumentum ad populum" is not an appropriate response. My 10,000 others seem to be quite happy was merely to point out that the poster was one of a small minority - I wasn't suggesting he was wrong! As for the main issue (Incidentally, the sentence However, he has not addressed the main issue of why the non-self-serving first-hand published reports of his fellow R-Pi forum administrators, including Broadcom employees JamesH and Dom, should be considered unreliable third-party reports, unsuitable for Wikipedia. seems to be counter to your own argument - I believe there is a NOT missing towards the end), on first reading the article, I realised the accuracy, verifiability or appropriateness of various sections was in need of improvement. I dealt with what I considered to be the most urgent requirment first. "Two wrongs don't make a right" is an expression I well remember from my childhood. It may not be commonly known in California, or wherever you are, but it is a well-used phrase where I come from. That I hadn't tackled the other issues I saw is a matter of time available, and other, non-wikipedia priorities. (and at the time, they weren't fellow administrators in any case. As I mentioned above, I was only asked to become an Admin after all this took place (and before you ask, No, it wasn't because of all this!).

Having been editing Wikipedia since 2005, and with several thousand edits behind me, I think I do have some idea of what is required, and would say that even forum entries from acknowledged RP Foundation members cannot necesarily be considered as a reliable source, unless perhaps made in some official capacity. As a single-interest editor (or perhaps a sockpuppet), I'm not sure what your credentials are, but I suggest it may be more appropriate and productive to take a less agressive attitude towards others with whom you do not agree.

I think I have made enough of my point. My conscience is clear. Regards, Lynbarn (talk) 20:11, 15 February 2012 (UTC)[reply]

You can demand to see my credentials on the R-Pi Forum, but not here. I slept at a Holiday Inn Express last nite. OK? You can slag off the posters on the R-Pi Forum, but not here. Liz is appalled that there are editors here that aren't R-Pi Forum members in good standing. So what? Liz doesn't make the rules here. I could ask about your credentials, but you've already boasted that you didn't understand a word of the R-Pi datasheet, so I won't bother. You raise the issues of accuracy, verifiability, and appropriateness. Those are reasonable issues. But it is very well established that compile speed is atrocious on the R-Pi, just as it is for similar ARM processors with limited memory and a slow swap device, for modest applications like Stellarium that JamesH reported takes 5 hours rather than the 5 minutes or less it's supposed to take. Can you imagine walking into your child's classroom and seeing all the kids with their heads on their desks waiting the 5 or so hours it takes to compile their code, and not being able to use their machines for anything else while they're waiting? No, I can't either. There's a good reason why Pentium II class computers were removed from classrooms years ago. 72.211.206.238 (talk) 06:58, 16 February 2012 (UTC)[reply]

72.211.206.238, put the flamethrower away. This is Wikipedia. You have two choices; you can choose to be WP:CIVIL or you will be blocked from editing Wikipedia.

Re: "It is very well established that compile speed is atrocious", either provide a citation to a reliable source that says exactly that (blogs and forum comments are not reliable sources) or stop making the claim.

Re: "Can you imagine walking into your child's classroom and seeing all the kids...", stop it. Just stop it. Nobody here is going to pay the slightest amount of attention to your speculation. We will consider that argument right after a teacher outfits a class with Raspberry Pis and reports - in a reliable source - the results. Nobody cares what you think will happen in a classroom. Even if you are right nobody cares. Even if your arguments make perfect sense nobody cares. Even if you buy a bunch Raspberry Pis, put them in a schoolroom and post a video showing you are right, nobody cares. We don't care about any of that because it none of that is a citation to a reliable source. So just stop it. You are wasting everybodies time with stuff that Wikipedia cannot use. --Guy Macon (talk) 10:11, 17 February 2012 (UTC)[reply]

Stop your crystal ball gazing, and dismissing the ability of the RasPI to provide a platform for kids to learn to program in Python before the official software package to do so is even released. These kids do not want to compile extremely large application programs, they want to run "hello world" programs, up-to maybe a hundred lines of code, on a software package expressly tailored to do that. If such a package is released and cannot do that, then you can add information about it on wikipedia. Wikipedia is not a soapbox for your POV. Mahjongg (talk) 09:54, 16 February 2012 (UTC)[reply]

I do believe that people either in the Raspberry Pi group or related interests have been trying to spin this article to be more positive for them. I think the most important thing that needs to be driven into people is: "Though the raspberry pi queen has 100% control over her blog doesn't mean her fanboys can do what they want on wikipedia". The raspberry pi is an interesting product and I'm sure it will be very popular, but it sure the heck isn't the "jesus" computer that you all act like it is. The ONLY reason that people care about it...is the price...and if you think otherwise then sell it for over $100 to see how many people truly think the same about it. • Sbmeirow • Talk • 11:14, 16 February 2012 (UTC)[reply]

I believe there is a fundamental difference between how two groups look at the RasPI. one group seems to think the RasPI is somehow a "PC for $35", and they compare it with a normal desktop computer, and point out its (obvious if you look at it that way) shortcomings. The "enthusiasts" see the potential for the RasPI to fundamentally change how "computer education" might change from "learning about IT" to going back to the fundamentals about having fun learning about how computers work on a fundamental level, and how to program them. Its what kids learned on the BBC Micro, writing simple little BBC BASIC programs. They see the rasPI as a "BBC Micro 2", bringing back what was good about learning to understand how computers work, by writing small programs on them. Obviously no longer by using BASIC, but by using a more modern but still appropriate language (such as Python). On which you have immediate and rewarding feedback, so its still fun to do. Obviously Wikipedia shouldn't be, (and isn't) a platform for "spinning" any story, whether negative or positive. If you look at it as if it was a "low cost PC replacement", you can only become disappointed, it certainly isn't a "jesus computer" (a terminology which seems to be borrowed from iPAD bashers). But that simply isn't what the RasPI is for, it is an extremely low cost (so massively applicable) platform to learn programming, and the fundamentals of how computers work, on. And as such I hope it will succeed. It isn't a "honda accord" it might not even be "road safe", but I hope it will be a great "tricycle" for kids learning about computer fundamentals. Mahjongg (talk) 23:41, 16 February 2012 (UTC)[reply]

Mahjongg, I don't think it makes sense to say small programs will work and large ones will not. It also matters what language features are used. I believe integers take 16-bytes each, so a simple loop like the following can take 160MiB. It simply prints the factors of x, that is, the numbers whose remainder is 0 when divided into x. Give it a try if you have python installed. And yes, I know this example can be made more efficient using xrange.

                         x = 9999999
                         for i in range(1, x):
                             if x % i == 0:
                                 print i

It also seems unlikely that R-Pi users will only compile small programs. It used to be the case that beginners would type in code from magazines, since it's much easier to modify someone else's code than to start from a blank screen. The equivalent these days is that students tend to download large amounts of source code for applications, games, and utility libraries from the internet, and expect to be able to compile it. 72.211.206.238 (talk) 07:00, 17 February 2012 (UTC)[reply]

You are still concentrating on the wrong "problem", and still assuming the RasPI will be like a PC, running on top of a typical PC OS that takes up 80% or more of the memory. I'm assuming instead that the RasPI will be like a BBC micro with BBC basic replaced by some form of a limited Python system, without large (GUI) libraries, and running on as simple an OS, with as small a memory footprint, as is possible, do not think "windows", but think DOS with a dedicated graphics library. It remains to be seen whether such a system will perform satisfactory, but I would not simply dismiss it beforehand, as you seem to want to do. Mahjongg (talk) 09:27, 17 February 2012 (UTC)[reply]

We need to edit the page so that is does not contain any hint of either assumption. Instead it should report what is in the sources, and if the sources don't exist (likely considering it hasn't shipped yet) we should be silent until they do. --Guy Macon (talk) 01:23, 18 February 2012 (UTC)[reply]

Agreed without reliable sources there is simply nothing to report, one way or the other. Until then we should remove all speculation. Mahjongg (talk) 15:55, 18 February 2012 (UTC)[reply]

Quote spam?

Edits in the 'Reception' section added quotations in the refs. These have since been removed, with the edit summary delete quote spam. The quotes were not included as spam, but to make the references more complete without readers necessarily having to follow the external links. In fact, WP:DEADREF encourages their inclusion: Search the web for quoted text or the article title. I therefore propose to restore the content. -- Trevj (talk) 14:01, 16 February 2012 (UTC)[reply]

I disagree, because some of the links had large amounts of text, thus polluting the reference section with tons of text they can't get into the body of the main article or toooooo lazy to rewrite and include in the article. If the quote was hidden and didn't show up on the reference section, then I wouldn't mind, but unfortunately the template doesn't hide the text. WP:DEADREF said SHORT quote, NOT multiple sentences like some people were doing in this article. Everyone knows how to follow links, so it really isn't needed. • Sbmeirow • Talk • 15:13, 16 February 2012 (UTC)[reply]

typically of me, I can see both sides here. The original quote was far too long, and if all quotes were like that, the reference section would be as long as the article. However, if WikiGuidelines encourage such quotes, then perhaps a single, short sentence could be included. Personally, I would leave it out. As Sbmeirow says, the link is there to be followed. Regards, Lynbarn (talk) 16:34, 16 February 2012 (UTC)[reply]

Thanks for the comments. This is a general issue, wider than this article alone. I intend to follow up the (optional) inclusion of {{Show}} within ref templates and will post here when I've done so. -- Trevj (talk) 15:53, 24 February 2012 (UTC)[reply]

third-party tag

This page currently has a third party tag, which results in this being displayed:

I have reviewed the page and, in my opinion, the tag is not justified. Rather, a product that is notable but has not yet shipped is naturally lacking in third-party reviews, simply because nobody has hardware to review. I propose removing the tag. Agree? Disagree? --Guy Macon (talk) 01:20, 18 February 2012 (UTC)[reply]

This is discussed at above, and some alternative tags suggested. I'm happy for it removed, as at least some of the issues I was highlighting hav enow been resolved. Regards, Lynbarn (talk) 01:27, 18 February 2012 (UTC)[reply]

OK, I am going to remove it. If anyone wants it there, please speak up and I will put it back while we discuss the issue. --Guy Macon (talk) 16:44, 18 February 2012 (UTC)[reply]

"The 26-pin GPIO connector is not assembled on the board."

Re: "The 26-pin GPIO connector is not assembled on the board", this is a decision than can be changed right up until the boards are assembled. Do we have some statement from the RPF that they definitely won't add the header? If so, do we have a definite statement that the holes will be left open and not filled with solder (leaving them open is slightly more expensive). --Guy Macon (talk) 16:43, 18 February 2012 (UTC)[reply]

It may be that nobody really knows: see THIS - the entry from Liz at 6:30 pm Jan 20th. regards, Lynbarn (talk) 16:54, 18 February 2012 (UTC)[reply]

Which says "Current guess is they will not be populated." I an going to remove the material as being speculation that violates WP:CRYSTAL. A lot of these details can be put back in to the article after the boards start shipping and reliable sources start describing what they got. --Guy Macon (talk) 18:17, 18 February 2012 (UTC)[reply]

Similar products

The section with the above title has two lists, thus:

BeagleBoard, Cotton Candy, CuBox, Hawkboard, IGEPv2, OmapZoom, PandaBoard, Pandora
Arduino, ARM express, BASIC Stamp

with no indication of what differentiates the two. Why are they laid out this way? Andy Mabbett (Pigsonthewing); Talk to Andy; Andy's edits 19:58, 24 February 2012 (UTC)[reply]

I divided them that way because the first list are all general purpose computers (at least to some degree, as far as I am aware - though I may have missed one or two). The Pi is similar to them in being a small, general purpose computer (runs a regular operating system, etc.)

The second list are microcontrollers with extra assistance intended to make them more accessible to the novice (e.g. programmed in BASIC, etc.) The Pi is similar to them in that it is targeted at novices/learning, and it would probably be usable in many of the same contexts as some microcontrollers. (An early idea of the Pi was a microcontroller that was programmed in python.)

So you can look at the Pi as a really cramped general purpose computer, or as a simple microcontroller with a lot of extra memory (and some special I/O). Zodon (talk) 09:16, 26 February 2012 (UTC)[reply]

I should've probably dropped a note here in conjunction with this edit, but I didn't spot this heading. Sorry. -- Trevj (talk) 22:14, 1 March 2012 (UTC)[reply]

Reception

Undue weight to fringe pessimism. Reception is overwhelmingly positive on the whole. Highlighting the witterings of these miserable so and sos is not a proper way to balance COI concerns. Needs more of the extensive positive press mentioned if these criticisms are to stay. Added template. Rubiscous (talk) 18:11, 26 February 2012 (UTC)[reply]

I have a problem with the reception section. My problem is that neither the linked citations or the section itself talk about the Raspberry Pi, but rather use the RP as a coatrack for editorials about how best to educate children. Yes, we have a rough balance between editorial opinions favoring or opposing teaching children programming with a Raspberry Pi, but no discussion about actual Raspberry Pi's.

(Irony / sarcasm = ON) Why, it's almost as if none of the editorials were written by someone who had actually held a Raspberry Pi in their hand! And it's almost as if the Raspberry Pi Wikipedia page was written be a bunch of editors who not only have never actually held a Raspberry Pi in their hand, but cannot find any sources other than the RP foundation written by anyone who has! (Irony / sarcasm = OFF)

As I have pointed out before, you really can't write a reception section before anyone has received a Raspberry Pi and written about it. Speculation doesn't count. Editorials about unrelated topics are right off. I propose that we delete the reception section and write a proper one once we know how the Raspberry Pi is received.

This reminds me of an old joke. The new wife of an IBM salesman goes to a judge wanting an annulment because he never consummated the marriage. She explains; "instead he just sits at the foot of the bed every night telling me how great it's going to be!" That's pretty much the situation with the Raspberry Pi. --Guy Macon (talk) 20:39, 26 February 2012 (UTC)[reply]

Today (Feb 29, 2012) it went "on sale." Two selected distributors don't have links for shipping to US. On one of those two site, once you get to the page with the board, there is only a form to "express the interest in the board" form. No buy button, nada. Both sites are overwhelmed with traffic (they call themselves "worldwide distributors"). According to the new text on the main RPi page, this is six years since project inception. — Preceding unsigned comment added by 67.164.110.16 (talk) 08:59, 29 February 2012 (UTC)[reply]

It won't be long before real feedback starts flooding the internet, because 10K of them are coming from production very soon. • Sbmeirow • Talk • 21:04, 26 February 2012 (UTC)[reply]

I'm going to use one to control my RepRap, because they'll both be here just about in time for each other. Andy Dingley (talk) 21:07, 26 February 2012 (UTC)[reply]

Not sure about the template - what ideas are being given undue weight? There is some pro and some con mentioned. I do not think it is premature to cover this. After all, part of the impact is what effect the concept has. (The $100 laptop had considerable impact before they ever released a machine.)

Likewise the conceptual model here may have elicited reaction, before we have hardware reviews. More coverage of the idea behind the Pi would be helpful to the article (whether in this section or elsewhere). Why they think these particular price points are important, why they think that this is important element in revitalizing interest in CS, etc. Zodon (talk) 01:49, 27 February 2012 (UTC)[reply]

We have one sentence that describes a pro opinion piece, one sentence that is mostly neutrally stating facts about school enquiries, a short weasel-worded sentence about Eben Upton's hopes that should be attributed to him, then a paragraph describing con opinion pieces. There's slightly more con. Not that pro and con should be equal: WP:UNDUE states "Wikipedia should not present a dispute as if a view held by a small minority deserved as much attention overall as the majority view. Views that are held by a tiny minority should not be represented except in articles devoted to those views". Is there any reason to believe these 'con' viewpoints are shared by anybody further than the bloggers who are expressing them? Same goes for the 'pro' viewpoints. We should have third-party sources that discuss these opinions having been made rather than taking them straight out of the horse's mouth, otherwise who's to say they're notable? Rubiscous (talk) 06:18, 27 February 2012 (UTC)[reply]

I think we should nuke the section. It shouldn't be long before we have a wealth of reviews from reliable sources, at which point we can recreate it. What is there now simply isn't helpful to the reader. --Guy Macon (talk) 07:33, 27 February 2012 (UTC)[reply]

I don't think it's fair or balanced to remove the section. A neutral point of view is representing fairly, proportionately, and as far as possible without bias [...] WP:CRITICISM (although I know that's not an actual policy) suggests use of a 'Reception' section for this. The {{Undue}} tag is fair enough, particularly because the section should represent views proportionately: even though the boards aren't (quite) available yet, the balance of reception pieces seems to be overwhelmingly positive.

Therefore IMO the answer to resolving the issue is to include more +ve viewpoints to balance the couple of -ve ones which have been published in reliable sources. (Which is what Rubiscous suggested above.) Removing the section is not warranted - and as Guy Macon says, it'll be recreated very soon anyway. -- Trevj (talk) 10:37, 27 February 2012 (UTC)[reply]

How can a website be considered a reliable source about a Raspberry Pi when they have never seen a Raspberry Pi and are only writing about what they think a Raspberry Pi will be like sometime in the future? --Guy Macon (talk) 08:25, 28 February 2012 (UTC)[reply]

Per WP:RSOPINION? I've added a further +ve comment and suggest the {{Undue}} tag now be removed. -- Trevj (talk) 09:17, 28 February 2012 (UTC)[reply]

I agree that the undue weight concerns have been addressed, so I boldly removed the tag. I also note that the reception section mentions The Centre for Computing History, which has one of the prototypes. This is a Good Thing; a report of actual reception to the board as opposed to what someone thinks the board is going to be like. These are some real improvements. Good work! --Guy Macon (talk) 17:54, 28 February 2012 (UTC)[reply]

Still have not shipped any product....

From [ http://www.raspberrypi.org/ ]:

"Although we are still waiting for units to arrive from China, you can start buying the Raspberry Pi today." --Guy Macon (talk) 09:13, 29 February 2012 (UTC)[reply]

That's because they've switched plans. The resellers have their own infrastructure for shipping to customers worldwide direct from China. They won't all come to the UK first now. Rubiscous (talk) 11:32, 29 February 2012 (UTC)[reply]

We need to watch for credible reports that boards are arriving at customer's homes/businesses. This is an important milestone which allows Wikipedia to say things (properly sourced, of course) that WP:CRYSTAL does not allow us to say before that happens. --Guy Macon (talk) 17:54, 29 February 2012 (UTC)[reply]

Currency

As this is a wholly British product, i.e. designed, produced and marketed in Britain, this article should use the pound sterling as the primary currency and US dollars for an international comparison in brackets next to the sterling. 81.129.9.198 (talk) 14:06, 29 February 2012 (UTC)[reply]

They're the official figures given by the foundation themselves. Note how their subtitle was "An ARM GNU/Linux box for $25. Take a byte!" -- 李博杰 | —Talk contribs email 14:49, 29 February 2012 (UTC)[reply]

Oppose This has been said before. The current version includes both currencies (with my addition at the time using the order you propose). The various news reports use different currencies, although the Foundation itself frequently refers to the $25 price point, which can be taken as an indication of its world view position. Therefore I consider the current primary currency of US dollars to be appropriate. -- Trevj (talk) 14:55, 29 February 2012 (UTC)[reply]

There is no need to vote, and your vote will not be counted. This Wikipedia page will list prices in USD followed by GBP in parenthesis until such time as the RPF starts listing the price in GBP. It is a core Wikipedia policy to follow the sources. --Guy Macon (talk) 22:03, 29 February 2012 (UTC)[reply]

Sorry for creating confusion, my "vote" was a !vote substantiated by the explanation. Anyway, by way of further explanation, per MOS:CURRENCY US$ should be used. And my understanding is that this should remain the case even if the Foundation choose to use GBP in their headline. -- Trevj (talk) 00:43, 1 March 2012 (UTC)[reply]

For The US currency mention by the RPF was in relation to the US market NOT international, it is plainly ludicrous to have a British product primarily listed in $ and looks like an attempt to 'possess' the product. For goodness sake it's not even available for sale in the USA yet. I have edited the currency to reflect the reality and hope I have done it in a neutral way as possible as we all want it to succeed Twobells (talk) 15:11, 29 February 2012 (UTC)[reply]

- Please, read response from benlisquare above. If you have problems with using USD, write an email to RPi's webmaster, and ask them to start endorsing pound sterlings. Otherwise, you are just being biased to your POV, not respecting the owners of the product. — Preceding unsigned comment added by 67.164.110.16 (talk) 16:44, 29 February 2012 (UTC)[reply]

There is no need to vote, and your vote will not be counted. This Wikipedia page will list prices in USD followed by GBP in parenthesis until such time as the RPF starts listing the price in GBP. The page must reflect what the sources say. --Guy Macon (talk) 22:03, 29 February 2012 (UTC)[reply]

I changed the page back to US dollars, followed by pounds sterling in parenthesis. If you wish to make pounds sterling the primary unit, you must first convince the Raspberry Pi Foundation to price it that way on their website and news releases. Also, all pounds sterling prices must be preceded by a "~". If you want to change this, must first convince the Raspberry Pi Foundation to fix the pounds sterling price and let the US dollar price price float as the currency exchange rate varies. This is standard Wikipedia policy; we must follow what is in the sources. Please note that there is no point voting "support" or "oppose"; following the sources is a core Wikipedia value and is not optional, nor is is subject to consensus. (BTW, my preference would be for both the RPF website and Wikipedia to use pounds, but that is not my decision to make). --Guy Macon (talk) 17:18, 29 February 2012 (UTC)[reply]

Oppose The price is in Dollars because the components it is made up of are priced in Dollars. They're running on such a tight margin that exchange rate fluctuations could make a fixed Sterling target price inviable. We should mention this in the article, should hopefully stop silly requests like the above. Rubiscous (talk) 17:34, 29 February 2012 (UTC)[reply]

There is no need to vote, and your vote will not be counted. This Wikipedia page will list prices in USD followed by GBP in parenthesis until such time as the RPF starts listing the price in GBP. Wikipedia must use whatever currency the RPF uses. --Guy Macon (talk) 22:03, 29 February 2012 (UTC)[reply]

Things that are inserted in articles should be for the benefit of the reader, not the editor. Besides, Wikipedia editors don't need to know why the RP is priced in USD. They only need to know that it is and that we must follow the sources. --Guy Macon (talk) 17:48, 29 February 2012 (UTC)[reply]

The fact that an editor was confused suggests that many readers would also be confused as to a British foundation's reasons for selling in Dollars. The reason I stated above is the official explanation. If sourced it should go in. What I described above is but a secondary benefit. Rubiscous (talk) 03:23, 1 March 2012 (UTC)[reply]

oppose: The price is fixed by The Foundation at $25/35. This has for a long time been the Foundation's stated policy. The UK price will be allowed to fluctuate with the exchange rate, which is the reason the UK price quoted by the distributors is £21.60 at present, not the previously estimated £22. Brent Crude is always priced in USD, why not the 'Pi? Regards, Lynbarn (talk) 20:44, 29 February 2012 (UTC)[reply]

There is no need to vote, and your vote will not be counted. This Wikipedia page will list prices in USD followed by GBP in parenthesis until such time as the RPF starts listing the price in GBP. See WP:RS and WP:V. --Guy Macon (talk) 22:03, 29 February 2012 (UTC)[reply]

oppose: The Foundation has ALWAYS stated the price in US dollars. • Sbmeirow • Talk • 21:36, 29 February 2012 (UTC)[reply]

There is no need to vote, and your vote will not be counted. This Wikipedia page will list prices in USD followed by GBP in parenthesis until such time as the RPF starts listing the price in GBP. PLEASE STOP POSTING "OPPOSE" OR "SUPPORT" VOTES. --Guy Macon (talk) 22:03, 29 February 2012 (UTC)[reply]

In fairness, Guy Macon, I don't think anybody believes they are voting here - merely expressing an opinion on the section's original comment :) Lynbarn (talk)

Slackware

"While Slackware can load and run a GUI, it was designed to be run from the shell." While running from the shell is certainly an option, as it is with any other Linux distribution, X is very much a central part of Slackware. From the Slackware homepage: "The long-awaited Firefox 4.0 web browser is included, the X Window System has been upgraded (and includes the open source nouveau driver for nVidia cards)." — Preceding unsigned comment added by 32.129.14.124 (talk) 00:04, 1 March 2012 (UTC)[reply]

From The Slackware Book (published by Slackware Inc.):

"Differences Compared to Other Linux Distributions:

There are a great number of differences between Slackware and other mainstream distributions such as Red Hat, Debian, and Ubuntu. Perhaps the greatest difference is the lack of 'hand-holding' that Slackware will do for the administrator. Many of those other distributions ship with custom graphical configuration tools for all manner of services. In many cases, these configuration tools are the preferred method of setting up applications on these systems and will overwrite any changes you make to the configuration files via other means. These tools often make it easy (or at least possible) for a rookie with no in-depth understanding of his system to setup basic services; however, they also make it difficult to do anything too out of the ordinary. In contrast, Slackware expects you, the system administrator, to do these tasks on your own."

If you install Slackware, this becomes obvious from the moment it boots to the command line and you have to type startx to load the GUI. It runs the GUI just fine - great as a matter of fact - and you can edit your config scripts (from the shell, naturally) to autoload the GUI, but most of the help files, newsgroup postings, tutorials, etc. pound home the fact that you are expected to administer Slackware from the command line.

Contrast this with Ubuntu, Redhat, Debian, etc. where you have to tweak your configuration to prevent it from booting to the GUI. --Guy Macon (talk) 00:44, 1 March 2012 (UTC)[reply]

Regulatory compliance?

Sold in the EU without a CE mark? Sold in the US without FCC part 15 compliance or UL listing? Marketed to children without any safety compliance testing whatsoever? Kids will be able to lick the lead solder on the board, sleep with it plugged in under their pillow and die in a tragic fire, plug it into a live network while touching the board with their wet hands and get electrocuted... — Preceding unsigned comment added by 157.234.254.24 (talk) 03:34, 1 March 2012 (UTC)[reply]

All of the above is speculation. We simply do not know what regulatory marks will be on the boards until someone actually receives one. Unless the boards turn out to be a lot different than what has been announced, they will be incapable of shocking anyone (five volts maximum) or setting a pillow on fire ("3.5 watts", but is that peak or average?) --Guy Macon (talk) 03:51, 1 March 2012 (UTC)[reply]

They haven't been marketed to children. They won't be marketed to children until they have cases. Rubiscous (talk) 04:37, 1 March 2012 (UTC)[reply]

Reading the wording of this anonymous poster, and his unrealistic "dangers", it seems he is being satirical/sarcastic. However regarding compliances he does have a point, I see no mentioning anywhere of any complicance to EMC regulations. Saftety regulations are a moot point, as no dangerous voltages are used, and the whole thing will be powered with only a few watts, but I'm not sure in how far EMC regulations apply to the raspberry PI (which isn't a consumer device). If any EMC testing has been done it should probably have been done by Broadcom as the SOC seems the sole major source for any EMF, (except for the ethernet & USB bridge chip) so its quite possible that the raspberry PI has an automatic complyance, based on being build from a compliant SOC. Mahjongg (talk) 17:14, 2 March 2012 (UTC)[reply]

Accepting orders vs delivering boards

Terminology: As of 29 February 2012, we can say that the RBF is "taking orders for" or "selling" the model B. Be careful that your wording does not imply that they are selling the Model A. Terms such as "delivered", "produced", "released" or "manufactured" should not be used until a reliable source confirms that at least one board has reached a customer. This should happen within a few days, depending on the shipping method the RPF chooses. --Guy Macon (talk) 03:39, 1 March 2012 (UTC)[reply]

Please add a link

This edit request has been answered. Set the |answered= or |ans= parameter to no to reactivate your request.

To Raspberry Pi computer in action (BBC). — Preceding unsigned comment added by 79.119.87.229 (talk)

Not done: The information at that external link is already in the article, so doesn't add anything. — Bility (talk) 19:46, 1 March 2012 (UTC)[reply]

Link Discussion

Perhaps some clarification would be in order on the criteria here. The pastebin item is even more inscrutable (and less of an obviously reliable source) than a blog. Speed, etc. comparisons for the Pi would be interesting, but even after studying it for a while I am still not even sure what the pastebin item even says. (Yes, I know what linpak is, but what are results for the Pi, what are for Atom - not at all clear.) Zodon (talk) 06:53, 29 February 2012 (UTC)[reply]

A collection of links is a good idea (providing people have time to subsequently include them in the article). However, the short summaries you recommend including may detract from the normal consensus-building approach, whereby [e]ditors usually reach consensus as a natural product of editing. After someone makes a change or addition to a page, others who read it can choose either to leave the page as it is or to change it.

Additionally, I do have objection to the "warning" given to not quote from articles. Whether to quote or not has no consensus either way, so it is therefore at the discretion of individual editors whether to include any quote (and what length is appropriate). -- Trevj (talk) 12:06, 29 February 2012 (UTC)[reply]

I stated it for a reason. On another wikipedia article talk section, people with one view were quoting large amounts of text from articles (1 to 3 paragraphs), and it got completely out of hand. They got mad when we started to merge all the posts to one talk section, then another editor finally deleted everything and laid down new guidelines to get things back under control. That mess is the reason behind my comment for this section. • Sbmeirow • Talk • 21:42, 29 February 2012 (UTC)[reply]

Sounds like that editor had consensus from other editors to act in that manner on that talk page after protracted disruption. Consensus there does not transfer here. You don't lay down the law without consensus. You don't own this talk page. Request whatever you like but please refrain from "warning" people until after they have violated specific policies or guidelines or been repeatedly disruptive. Rubiscous (talk) 04:28, 1 March 2012 (UTC)[reply]

If you showed up to the party sooner, then you would understand why a few of us are trying to get this drunken brawl under control. At this point in this mess, putting up warnings in advance seems to work a lot better than spanking everyone later (or asking admins to spank people later). • Sbmeirow • Talk • 08:34, 1 March 2012 (UTC)[reply]

Linux kernel versus Linux OS based

The Raspberry PI is supported with a Linux kernel including all the drivers that are needed to get the most out of the hardware, that is terrific because for some parts of the hardware this will in fact be the only route to using the hardware, especially when the GPU is involved, but it seems also to be true for driving such low level hardware functions as the SPI and I2C hardware. This because of the closed nature of some parts of the hardware. Still this doesn't mean that this kernel and its drivers must be used in combination with a full (GNU)/Linux desktop OS! Its also very possible to built BBC BASIC on top of this kernel (to mention just one option), or any other operation system or programming environment.

This notwithstanding that as it seems the raspberry PI will come out with its own Linux distro.

This situation should be reflected in the article, but at the moment it is not! Mahjongg (talk) 00:18, 3 February 2012 (UTC)[reply]

Nor should it be. Please read the section titled "The time has come to put the Crystal Ball into storage", put away the crystal ball, and only report things that have already happened or are 100% certain to happen. --Guy Macon (talk) 02:09, 3 February 2012 (UTC)[reply]

Okay, fair enough. Mahjongg (talk) 09:51, 3 February 2012 (UTC)[reply]

Re: OS versus kernel. See: "Distribution"

Please note that the desktops used with Linux: LXDE, KDE, GNOME, and the many others, are not part the operating system at all. (Please don't conflate Microsoft's marketing terminology with what an OS actually is, http://en.wikipedia.org/wiki/Operating_system [6] "A collection of routines which directly manage hardware resources".) Remember that X-Windows is an OS Independent utility that runs on many different Operating systems, and hardware platforms, which was its design goal. The Linux Desktops that run on top of X-Windows are NOT part of the Linux operating system. They ARE Desktops, but they could be run on any machine that runs X-windows, and some are, for example SUN's, MIPS, MAC's etc... You could even see them on VMS systems (if you really, really wanted to. ewww. :-) )

I don't want to get into exactly where the line between the OS and "Not part of the OS" is because that will devolve into a long and likely, unresolved discussion, but suffice it to say, no desktop which runs on top of X-windows is part of the OS on any operating system. The whole design of X-lib, which X-Windows runs on top of, was to make certain that X-windows was totally decoupled from each OS it runs on. Both models of the Rasberry Pi are running the Linux OS, Not just the kernel.

The terminology that trips up many people when talking about Linux is the confusion between what an Operating system is, and what a "Distribution" is. X-Windows comes with some Linux Distributions and doesn't come with other Linux "Distros". And yet both types of distro are using the Linux OS, despite the second type not having a GUI desktop.

Example issue: Is the command line shell part of the Operating System in Linux? Well that depends, Which shell do you mean? Bash, Dash, Ash, Csh, rsh, tcsh, zsh, jsh, or ksh? (you would mean "sh" if you're a traditionalist.... ;-)

Discuss amongst yourselves. ;-)

Jjk (talk) 22:23, 1 March 2012 (UTC)[reply]

Let me explain what I meant before we go off on a wholly different tangent here. The idea that I actually meant to carry is that, perhaps surprisingly, the OS that you built on top of the piece of operating kernel software that is known as the Linux kernel does not have to be a traditional linux distro as we all know! You know a Linux kernel with GNU tools such as one of the traditional command line shells, with x-windows for graphics, and some kind of desktop manager running on top of that. Instead you can take the kernel and graft any other kind of OS on top, as long as it can hook into the kernel, and use the internal (unix like) structures. So in theory you can take the kernel and graft, lets say, RISCOS on top of it. Yes the result will be a hybrid, using Unix file mechanisms and message passing and such, and not the native RISCOS mechanisms, but its just an example. Another possibility is to use the kernel with a BBC BASIC like piece of software on top of the kernel. People often forgot that old 8-bit home computers (Including the BBC computer) also had kernels,as part of their BASIC in ROM. The one used in several Commodore home-computers from the 80's was even called KERNAL (not a misspelling), and MSX systems also had a rudimentary kernel in their ROM, that could even be expanded with disk routines in an expansion ROM. Such kernels were extremely primitive compared to the linux kernel, but it would be possible to take the Linux kernel and build something around it not even remotely resembling a traditional Linux (OS) distro, and maybe having a much smaller footprint. For dedicated applications, such as a dedicated "Python based machine", or call it a "BBC 2". I'm not saying this should find its way into the article at the time, as Guy Macon rightfully reminds us that we should not place material in the article based on speculation. Still just saying that the system will only run LINUX (OS) because only through the linux kernel you will have access to all the hardware functionality is also speculation. Other things might be done with this kernel which is explicitly modified/written for the hardware of the Raspberry PI to unlock its potential. The Raspberry PI is a system with limited memory, so it makes sense to not build things on top of a complete Linux OS, but directly on top of the kernel instead, forgoing even a single command shell, not even "sh". ;-) Mahjongg (talk) 03:11, 2 March 2012 (UTC)[reply]

I agree fully with the above, keeping in mind that all of this is speculation (even though we all know it is going to have a Linux Kernel, all we can do now is talk about "...plans to..."). I wouldn't be surprised if we don't end up with users having multiple SD cards, one that boots to BBC Basic, another that boots to FORTH, one with a Full Linux Distro, etc. The good news is that we don't have to guess; all we have to do is wait and report on whatever the sources say about what ships with the RP as well as any notable alternatives. --Guy Macon (talk) 09:20, 2 March 2012 (UTC)[reply]

Exactly, I for one am glad I don't have to speculate whatever there might be developed for the Raspberry PI, on top of the Linux kernel, (or relevant parts of it) because nobody can guess what the community will come up with next. What I can predict is that an average Raspberry PI user will probably have a small collection of SD cards, one for each purpose. So that (for example) after haveing studied Python for a bit, he can use another card to boot his favorite game. Mahjongg (talk) 16:09, 2 March 2012 (UTC)[reply]

Last night I was at a friend's house and he told me he stayed up half the night trying and was finally ably to buy a Raspberry Pi. He says they estimate delivery to California around March 16th. I won't be surprised if UK orders arrive quicker than that. --Guy Macon (talk) 21:42, 2 March 2012 (UTC)[reply]

"This board is intended to run operating systems based on the Linux kernel, including, but not exclusive to Linux[2] and to support the Python programming language,[11][12] BBC BASIC,[13] C[11] and Perl.[11]" This first half of this sentence could do with a re-write as it's unclear what it's saying "based on the Linux kernel, including, but not exclusive to Linux and to.." Including what? Are we supposed to have some examples here (like Debian & Fedora that have spins for it). What isn't exclusive to Linux, the device itself? How about: "This board is intended to run operating systems based on Linux, and to support the Python programming language, BBC BASIC, C, and Perl; however it may be compatible with other operating systems" or some-such? Twirrim (talk) 22:00, 2 March 2012 (UTC)[reply]

NO! haven't you read the above? The reason it is worded that way is exacly to avoid using your suggestion! What is the impetus that everything must run on top of a (full) Linux (distibution)? Doing that takes a very large portion of the small amount of memory there is away! If you are only interested in running Quake (or a similar game) then why run it on top of a full desktop system? The kernel has built in openGL-ES support so a game can drive the 3D engine directly. Its like the original x-box which ran a OS derived from the kernel of Windows and direct-X, but did you ever see a windows desktop? Do you know FORTH? If you knew it you would know that FORTH is its own "OS", it doesn't need to run on a desktop (although truth be told it often is used that way, when not running on a microcontroller)! In fact it wouldn't need a desktop to run on a Raspberry PI either, except for the fact that the only way to access essential functions of the PI's hardware is through drivers incorporated into this specially written version of a Linux kernel. so even FORTH has to use the kernel! That doesn't mean you have to use the other 80% of the rest of memory eating stuff that makes up a full desktop system. Running a complete Linux distro (whichever one it might be) can have its merits sometimes, and it will certainly be the case that much software will be written for such a combination, but certainly not for all possible applications will that be the optimal solution, and I cannot image that it will take a long time for software developers to figure that out. Mahjongg (talk) 01:32, 3 March 2012 (UTC)[reply]

I really don't care about the pissing about over what is and what isn't the kernel, OS, what is or isn't necessary or whatever that you're obsessed with. That's not something that has any place impacting this article. This is speaking as a sysadmin who gets pissed off at finding full GUIs installed on servers (let alone lightweight systems like this) and who spends his days looking at shell prompts. I totally agree with where you're coming from on that front. Give me the bare minimum necessary and I'm happy. What you're leaving behind in the article though is an ugly nearly nonsensical paragraph that will only confuse readers. You're not accepting any other alternative phrasing so pick your own, but it really needs changed. The wiki guidelines are quite clear: "Be plain, direct, unambiguous, and specific." "be as concise as possible—but no more concise". By aiming for conciseness what we've got is unclear, and confusing, let alone the fact that it doesn't make grammatical sense. The majority of people who read this article do not care and won't know about the absolute precise definition of what is and what isn't an OS. If you want to keep it concise and not risk giving the impression that a full OS stack is necessary: "This board is intended to run operating systems based on the Linux kernel, and to support the Python programming language, BBC BASIC, C and Perl." That's absolutely true to what the Raspberry Pi foundation have stated all along, and cuts out the bad grammar. For what it's worth the latter part of that sentence could do with a little shuffle around along the lines of "support the Python, BBC BASIC, C and Perl programming languages" otherwise it's ambiguous what BBC BASIC, C and Perl are.Twirrim (talk) 09:44, 4 March 2012 (UTC)[reply]

To Mahjongg: Gently, gently... More Light and Less Heat. All caps and exclamation points are not the way. If someone appears to have missed the point; assume good-faith, assume that you must have not explained it well enough and patiently explain it again. At least the first time they don't get it - eventually WP:IDIDNTHEARTHAT applies. A calm, cool dispassionate attitude and arguments based upon evidence (especially references to Wikipedia's policies and guidelines) and logic works best.

To Twirrim: Your response did not meet Wikipedia's standards for civility as explained in WP:CIVIL. In particular, calling another editor's legitimate objections "pissing" and "whatever that you're obsessed with" is unacceptable behavior. Please apologize to Mahjongg and continue making your argument without further incivility.

As for the issue at hand: Clarity is good. Brevity is good. confusing Linux with the Linux Kernel is not. This is an encyclopedia, and the questions "what is and what isn't the kernel" and "what is and what isn't the OS" are vitally important to get right. Suggestions for improving the clarity and brevity of the passage we are discussing are more than welcome -- I agree that it is a bit awkward and possibly even unclear as it is now -- but any such suggestion must retain the important distinction between Linux and Linux Kernel.

A related point (not implying that either one of you got this wrong, just that some people do) is that the following four things are completely separate and distinct on a Raspberry Pi:

Kernel (one specific instance: Linux kernel)

Operating system (one specific instance: Linux)

Windowing system (one specific instance: X Window System)

Desktop environment (one specific instance: Openbox)

Notes: Windowing system + Desktop environment = Graphical user interface. Microsoft combines all of the above and calls it the "Microsoft Windows Operating System"

On a Raspberry Pi, any of the above components are replaceable. We really cannot even assume that someone won't adapt BSD, QNX or FORTH to run on the Raspberry Pi. --Guy Macon (talk) 14:59, 4 March 2012 (UTC)[reply]

Ok Guy, I apologize for the inappropriate uppercased "O" and the acclimation point, ;-) And also to Twirrim for coming on so strong. I tend to become slightly exasperated as people so totally miss the point.

The only thing I want to add to Guy's comments is that it seems at the moment that to code anything at all for the Raspberry PI you seem to must use the specific kernel written for the Raspberry PI, simply because that is the only available code to access the graphical capabilities, (through the OpenGL driver built into the kernel, providing interfaces to the commercial non open "software blob" that drives the video logic) and maybe some other "black box code" for booting, USB access, and such. Perhaps its possible to lift this code (and the "blob") from the kernel, and re-use it in another kernel such as the one for RiscOS, but I'm unsure how practical a proposal that is, so for the moment it looks like the code has to be "married" to this kernel. On the other hand several sources say that RiscOS works, (so it drives the video logic) but only lacks USB drivers. So does that version use the Linux kernel? Or has it somehow succeeded to drive the video blob without needing the kernel. It will be interesting to see how all this will work out.

By the way, with the power of an older Xbox, but completely open, I wonder if anybody realizes what influence the PI might have as a new games platform. For non-commercial games that is, as I do not see how games can be sold for the PI. Except perhaps if you bundle such a game with all the logic of a PI (which after all is publicly available, if you can source the Broadcom chip) in a suitable hand held controller, somewhat like the re-make games on the market, like the C64 Direct-to-TV. Perhaps there is a market for a "Halo Direct-to-TV" box for $50?! Intel makes Flash chips with an interface that is the same as an SD-Card so it doesn't have to contain a discreet SD-Card, and you can epoxy seal the Broadcom chip and Intel Flash chip together.

By the way, English isn't my first language, so I'm always happy if someone makes my sentences more legible, but I'm not happy if they in the process remove exactly the distinction I wanted to make. So if you can change my wording without removing the distinction from using only the kernel, versus also using all the layers above which make up a typical linux OS system then please do. Mahjongg (talk) 16:29, 4 March 2012 (UTC)[reply]

Re: has [RISC OS] somehow succeeded to drive the video blob without needing the kernel. This is obviously WP:OR, but AFAIK yes this is the case. RO doesn't use the Linux kernel. Anyway, a few RO users have successfully placed orders, so once the port is made available I'm sure it'll be reported somewhere. -- Trevj (talk) 13:13, 5 March 2012 (UTC)[reply]

Yes obviously it would be WP:OR to put such a claim into the article, luckily no such restrictions are placed on a talk page, except that discussions should be used to improve the article. These are exciting days, we just have to wait and see what will transpire. I'm not convinced that RiscOS for the RPi doesn't use (part of) the kernel, I do not see how they otherwise can get any video out of the non documented video blob, except maybe if the blob (which also manages BIOS like tasks like booting etc.) somehow turns on a simple framebuffer mode they use, perhaps an emulated VESA mode? Did you read somewhere that they did it without using kernel code, or are you just assuming this is the case because they do not talk about it. I cannot find proof either way. Mahjongg (talk) 18:16, 5 March 2012 (UTC)[reply]

It cannot be true that the video blob is "undocumented." It has to be documented in the sense that you send it some sort of documented high level commands and out pops video. Anything that the Linux Kernel and/or X Windows can tell it to do is by definition documented, and thus some other OS can tell it to do the same things. The internals are no doubt not only undocumented but trade secrets, but neither RiscOS of the Linux Kernel needs to know the internals.

In fact, according to http://www.broadcom.com/products/BCM2835, the Broadcom BCM2835 system on a chip supports (I assume through the blob) OpenGL-ES. That's quite a bit more sophisticated than a simple framebuffer. --Guy Macon (talk) 20:29, 5 March 2012 (UTC)[reply]

Obviously the blob is documented, otherwise the writer who wrote the OpenGL ES driver couldn't have written it. Problem is more or less that the person who wrote the interfaces of the kernel drivers (between the blob and OpenGL ES, and probably also the interface code between the blob and X-windows) wrote it based on an NDA he signed to get info about the interface to the blob. I have heard that this interface code is so complex that reverse engineering it (to use the blob without OpenGL ES and/or x-windows) would be very hard if not impossible, so in principle the only way to use the blob is still -through- the use (interface) of the OpenGL ES driver and or x-windows, and these drivers are imbedded in the kernel. I wonder how x-windows access to 2D drawing (drawing the desktop) is done, directly to the blob, or somehow also through the OpenGL ES driver. In any case, as far as I know -all- video access (2D and 3D) goes through the proprietary code (blob) via undocumented (without an NDA) interfaces.

What I was supposing that RiscOS did was something completely different. Based on what I have heard, the "blob" doesn't only do video! The "blob" also boots the system. It bootstraps code from the SD-card with its own small RISC processor (not the main ARM, but a separate very simple RISC processor embedded in the GPU). This processor is the -only- one in the SoC which has a very small bit of permanent memory (ROM), which gives it just enough "smarts" to load the "blob code" from the SD card into memory and then turns on the real ARM, and with it executes the blob. The "blob" then acts like the BIOS in a conventional PC, and does all the tasks a normal BIOS does, for example it also initializes the video processor (GPU), and loads a font-set into RAM so a text mode can be supported (just like a conventional (video)BIOS does when initializing the standard VGA logic, and filling font RAM. It then does a second stage boot, and loads boot code from the SD-Card, which boots the actual OS. This code, (like in a PC) now immediately has access to a text mode display (for all the boot messages, etc). This is the point where either a linux kernel, or something else is booted. What my assumption was that from this text mode you can switch to an APA (highres) mode, using conventional "BIOS calls", the same way a VESA driver works, only these calls are not handled by a conventional (video)BIOS but by the blob code.

After switching on this frame buffer mode, the "blob" code doesn't have to be used at all, you can just write into the designated video RAM area, and its contents are displayed.

This means you can indeed have a APA frame-buffer, but not the accelerated 2D graphics the blob probably supports with a full x windows system, (nor obviously 3D support).

In conclusion: When using the kernel you can have accelerated 2D and 3D, without it you are stuck with a simple VESA frame-buffer system.

But, like I said, this is just my assumption of how things work. I would be happy to stand corrected by someone who knows exact details. maybe I am completely wrong in some respects. Mahjongg (talk) 23:15, 5 March 2012 (UTC)[reply]

GPU Maximum resolution

This edit request has been answered. Set the |answered= or |ans= parameter to no to reactivate your request.

Under "Specification", at "GPU", after "high-profile decoder", please add a new line with the following text:

   Maximum resolution: 1920x1080

to avoid disappointment for the 30" monitor crowd. :-)

I can look it up myself, but do you have a source for this factoid? Mahjongg (talk) 16:16, 2 March 2012 (UTC)[reply]

Well, its not mentioned in the FAQ (which is down at the moment, but a copy of it can be found here: [7]). So where is your source? I need a reliable source for this information to be able to add it. Mahjongg (talk) 17:00, 2 March 2012 (UTC)[reply]

Bit of a dead end with the manufacturer: BCM2835. --Ds13 (talk) 17:08, 2 March 2012 (UTC)[reply]

More on the BCM2835 here. But need datasheet for the VideoCore IV. --Ds13 (talk) 17:42, 2 March 2012 (UTC)[reply]

Yes, the facts about the details are a bit sketchy at the moment, for example we do not know (at least I could not find it) what the level of HDMI supported is, as HDMI 1.0..1.2a alone limits the resolution to 1920×1200. The default Video RAM size is (according to newly released information) 64MB, at 24 bits (3 bytes) per pixel (the lowest color depth supported by HDMI) this could then theoretically support a single frame buffer of 21 ⅓ megapixel. Comparatively a 1920 x 1200 resolution needs only 2,3 Megapixels. Even a framebuffer for the highest resolution supported by the latest version of HDMI (1.4) at 4096×2160 pixels uses no more than a fraction of the memory size available in the Raspberry PI. Obviously any modern video system will use multiple framebuffers (for double buffering etc,) and 3D displays will use massively more memory, but that wasn't the question. In lieu of a reliable source we should avoid mentioniing a "Maximum resolution". we will know the details soon enough when the first raspi's fall in the hands of the community. Mahjongg (talk) 17:48, 2 March 2012 (UTC)[reply]

According to the Quick user guide here:[8], the Raspi supports HDMI 1.3 en 1.4. Unclear is whether that means that the maximum supported resulution is 4096 x 2160 pixels, but it might be. Ill add the HDMI 1.3 and 1.4 info to the article, as its reliable sourced. Mahjongg (talk) 18:00, 2 March 2012 (UTC)[reply]

The element14.com Quick User Guide is a PDF created from this source: [ http://elinux.org/RPi_Hardware_Basic_Setup ]. --Guy Macon (talk) 20:29, 2 March 2012 (UTC)[reply]

Well, http://en.wikipedia.org/wiki/VideoCore makes that claim, but that's not authoritive of course. And just as well that you asked! Because [9] and [10] (currently down, read via Google cache) have RP project member "dom" apparently claiming that 1920x1200 is the maximum, not 1920x1080. (And he suggests that higher resolutions might be possible with advanced technical hackery, lower refresh rate, performance loss). The confusion perhaps stems from the fact that 1920x1080 is highest in the *TV* resolution list. Given those links, it think it's safe to write a text like "Maximum resolution: 1920 x 1200 (in the default configuration)". Although it's probably best to give it its own row in the table then, not in the GPU row. — Preceding unsigned comment added by 83.86.0.86 (talk) 18:06, 2 March 2012 (UTC)[reply]

Great! The Raspberry PI main site seems to be up again! Mahjongg (talk) 01:40, 3 March 2012 (UTC)[reply]

I've deactivated the template since this seems to be getting taken care of. Thanks, Celestra (talk) 19:33, 2 March 2012 (UTC)[reply]

Third video output option DSI

according to this: [11] there is a third video output option after composite and HDMI, namely Display Serial Interface (DSI) via the unpopulated 15 way flex cable connector. This might open extra possibilities, in particular the option to connect LCD panels. It seems it's using the same connector to which they planned to attach the smartphone style camera module which the RasPI foundation) plans to release next year. Mahjongg (talk) 05:33, 3 March 2012 (UTC)[reply]

Also, MIPI Camera interface. 91.154.87.201 (talk) 14:16, 6 March 2012 (UTC)[reply]

Links

Since the Raspberry Pi will be shipping soon, I decided to create this section, so we can post links here to related articles that might be useful for updating this article, either by validating or disproving something. WARNING: Do not abuse this section! • Sbmeirow • Talk • 23:01, 28 February 2012 (UTC)[reply]

Links to Forums

http://raspberrymod.com a Forum for discussions and HowTos — Preceding unsigned comment added by 217.6.106.73 (talk) 10:27, 1 March 2012 (UTC)[reply]

Links to sources that have not tested any hardware or which have only tested prototypes

http://elinux.org/RaspberryPiPerformance Raspberry Pi Speed Tests http://pastebin.com/yJj8WX3M (speed of Atom N270 @ 630 MHz (manually underclocked) vs Raspberry Pi ARM11 @ 700 MHz) • Sbmeirow • Talk • 23:01, 28 February 2012 (UTC)[reply]

http://boingboing.net/2012/02/29/raspberry-pi-launch-so-popular.html - Raspberry Pi launch so popular that retail partners collapse under load --Guy Macon (talk) 08:45, 29 February 2012 (UTC)[reply]

http://www.bbc.co.uk/news/technology-17190334 - Raspberry Pi computer in action --Guy Macon (talk) 08:59, 29 February 2012 (UTC)[reply]

http://gizmodo.com/5889245/five-things-you-can-do-with-the-new-raspberry-pi

http://www.thebackshed.com/forum/forum_posts.asp?TID=4748&PN=1 Critical / sceptical comments from Olimex --Guy Macon (talk) 19:14, 4 March 2012 (UTC)[reply]

Links to sources that have tested shipping hardware

(None so far; this should change soon) --Guy Macon (talk) 22:37, 29 February 2012 (UTC)[reply]

American English?

Why has this article been tagged with American English? Being a British product it makes far more sense to be written in British English. If nobody challenges this I will change it to British English SaintDaveUK (talk) 22:08, 4 March 2012 (UTC)[reply]

I agree, it is a little odd - perhaps the article was originally started by an American editor? It shouldn't be too difficult a job to convert it, and if you're willing to do so, I would support you in that. Regards, Lynbarn (talk) 22:14, 4 March 2012 (UTC)[reply]

American English, if it is consistently used, should probably continue. I think the Wikipedia Manual of Style is pretty clear about this: "When an English variety's consistent usage has been established in an article, it is maintained in the absence of consensus to the contrary. With few exceptions (e.g. when a topic has strong national ties or a term/spelling carries less ambiguity), there is no valid reason for such a change." The Raspberry Pi is not a strongly British topic (unlike examples given in the guideline). It is clearly an international product, and the Raspberry Pi Foundation itself chooses to express its pricing in US Dollars. The guideline also clearly states "An article should not be edited or renamed simply to switch from one valid use of English to another." Disclosure: I am Canadian and would probably favour British spellings, but I think it's worthwhile for editors to at least consider this guideline. --Ds13 (talk) 22:53, 4 March 2012 (UTC)[reply]

Considering the prduct's international visibility, I don't think WP:TIES is appropriate. WP:RETAIN states When no English variety has been established and discussion cannot resolve the issue, the variety used in the first non-stub revision is considered the default. If no English variety was used consistently, the tie is broken by the first post-stub contributor to introduce text written in a particular English variety. This may have been the introduction of the term "charitable organization". Now, "organization" is not my personal preferred spelling, but it is the spelling used in the non-redirected charitable organization article. For the record, I'm not personally fussed either way. (However, an article on the Raspberry Pi Foundation itself would IMO have those strong ties per WP:TIES.) -- Trevj (talk) 13:31, 5 March 2012 (UTC)[reply]

Guardian video review

See [12]. Worth adding a link? AndyTheGrump (talk) 18:04, 5 March 2012 (UTC)[reply]

Well, its almost "a review by the first customer who bought one". An event we have been waiting for to see happen. Except this isn't it, as its obvious he is not a "customer who bought one." , but perhaps it does count as Links to sources that have tested shipping hardware. Mahjongg (talk) 22:14, 5 March 2012 (UTC)[reply]

No - apparently it is a beta board :( [13] AndyTheGrump (talk) 22:20, 5 March 2012 (UTC)[reply]

Criteria for Inclusion as a Related Product

I would like to open a discussion as the what the consensus is for inclusion in the similar products section. I am going to list a few features. here are some silly / trivial examples showing the kind of answers that I think would be useful. --Guy Macon (talk) 21:12, 5 March 2012 (UTC)[reply]

SAMPLE ATTRIBUTES:

MADE BY THE RASPBERRY PI FOUNDATION?

Qualifies. Until they branch out into making toasters, anything they make is a related product.

MADE OF 100% MEAT?

Disqualifies. No product made of 100% meat belongs on the list of related products, no matter what other attributes it has.

ELECTRICAL?

Necessary. (also known as Necessary, but not Sufficient). Everything on the related products list must be powered by electricity, but being powered by electricity does not mean it belongs on the list.

BLUE?

Unnecessary. The color does not matter one way or the other.

OK, here are the actual attrib x64utes (feel free to add more):

ATTRIBUTES FOR RELATED PRODUCTS:

PROGRAMMABLE?

ARM BASED?

CASE, BUILT-IN KEYPAD AND SCREEN?

MEANT FOR EDUCATION?

OPEN SOURCE HARDWARE?

$100 OR LESS?

$1000 OR LESS?

NOTABLE ENOUGH TO HAVE A WIKIPEDIA PAGE?

CURRENTLY AVAILABLE (NOT DISCONTINUED OR FUTURE)?

EXTERNAL / REMOVABLE STORAGE?

I/O CONNECTOR(S)?

--Guy Macon (talk) 21:12, 5 March 2012 (UTC)[reply]

—————

Open source is a must. So is "Cheap", so $100 is an absolute maximum.

Don't see anything wrong with it being "future"...

Ytic nam (talk) 22:01, 5 March 2012 (UTC)[reply]

—————

I have Just two criteria for inclusion as a "similar product", except for that they also must be notable:

New product, Exclusively designed to help children learn to program, or to learn how a computer works

or

Very similar hardware, and price, that means all of the following:

ARM(6) + GPU based (GPU able to do 3D rendering etc)
256MB of RAM, maximum 512MB
very small, say max 10 x 10 cm (4 x 4")
Not a closed design (i.e. anybody may in principle make a copy of it, or use it as a basis for another product)
price lower than $50

Mahjongg (talk) 22:09, 5 March 2012 (UTC)[reply]

I'd say that 'Notable enough to have a Wikipedia page' is a must, regardless of other criteria - otherwise, this is an invitation to link-spammers. AndyTheGrump (talk) 22:14, 5 March 2012 (UTC)[reply]

Right you are, I modified my criteria accordingly Mahjongg (talk) 11:08, 6 March 2012 (UTC)[reply]

Unless I'm missing something, Similar Products and Related Products are not the same thing. Similar Products is currently along the lines of See also, i.e. wikilinks to articles. As the project matures there are certain to be numerous related products reported in reliable sources. I'd say that all such products can be included within prose in accordance with usual policies such as WP:UNDUE etc. -- Trevj (talk) 22:47, 5 March 2012 (UTC)[reply]

Well as I see it, the RPi is bound to be an "inspiration" for similar products or even regular clones, if that happens we need a section describing these. Also RPi itself has been inspired by earlier efforts, specifically OCPC en the Aasha tablet, so these should be mentioned too. Mahjongg (talk) 11:12, 6 March 2012 (UTC)[reply]

Use by activists

Raspberry Pi gets a mention in a news post [14] AngelFire3423 (talk) 13:10, 6 March 2012 (UTC)[reply]

[1] [1]

[2] [2]

[3] [3]

[1]

[2]

[3]

@@ Line 566: / Line 566: @@
 == Third video output option DSI ==
 according to this: [http://elinux.org/Rpi_Screens] there is a third video output option after composite and HDMI, namely [[Display Serial Interface]] (DSI)  via the unpopulated 15 way flex cable connector. This might open extra possibilities, in particular the option to connect LCD panels. It seems it's using the same connector to which they planned to attach the smartphone style camera module which the RasPI foundation)  plans to release next year.  [[User:Mahjongg|Mahjongg]] ([[User talk:Mahjongg|talk]]) 05:33, 3 March 2012 (UTC)
+Also, [[MIPI]] [[Camera interface]]. [[Special:Contributions/91.154.87.201|91.154.87.201]] ([[User talk:91.154.87.201|talk]]) 14:16, 6 March 2012 (UTC)
 ==Links==