Talk:Open Database Connectivity
|WikiProject Computing||(Rated B-class, Mid-importance)|
|WikiProject Databases / Computer science||(Rated B-class, High-importance)|
Error in a Paragraph
Under "Overview", the paragraph, "Despite the benefits of ubiquitous connectivity and platform-independence, ODBC has certain drawbacks. Administering a large number of client machines can involve a diversity of drivers and DLLs. This complexity can increase system administration overhead. Large organizations with thousands of PCs have often turned to ODBC server technology to simplify the administration problem." contradicts itself. It states that ODBC has certain drawbacks, leads into a large number of clients being a problem, and then states that businesses switch to ODBC to simplify the problem. It does not state any drawbacks of ODBC at all and instead, states a reason for switching.
Reference to SQL/CLI standard
ODBC is pretty much the same as the Call-Level Interface (CLI) as it is standardized by ISO in ISO/IEC 9075-3:2003.
Craig Stuntz reverted the changes made by User:220.127.116.11. I agree that the change was a bit suspicious. A drive-by-edit, by an anoynmous user, with no edit comments, might have been link-spam (but AGF), and lacked anything like a real citation. However, I'm not sure it is completely bogus. According to Simba's corporate history page, ODBC was originally a Simba product, which Microsoft licensed. If so, that's legitimate information, and makes the current article incorrect (the article says Microsoft created ODBC). More fact checking is needed, here. --DragonHawk 14:03, 14 July 2006 (UTC)
- Um, I did check the facts, and that's why I reverted. It is true that Simba was involved in development of ODBC. It is a misrepresentation, however, to imply that ODBC was wholly a Simba product, was developed by Simba prior to Microsoft, or that Simba was the only company involved. Note that the Simba page you link does not actually say what DragonHawk states above (the "Simba technology" referenced is a driver, not ODBC as a whole, and you don't need their SDK to make a driver), but seems crafted to make one think that it means something along those lines. The ODBC Hall of Fame, already linked in the article, gives a much more balanced presentation. As I noted in my edit summary, I reverted not strictly due to the linkspam, but because the edits were POV/misleading.
- We could discuss all the companies involved in creating ODBC, but (1) the passage in question is rather short and is accurate as it is and (2) we already link to the list I cite in the paragraph above. I don't really have any objection to a more balanced presentation of who was involved, but I don't think the article is hurting for lacking it, either. If you'd like to note what all of the contributors did (something along the lines of the Hall of Fame, for example), I have no objection. --Craig Stuntz 15:57, 14 July 2006 (UTC)
- Well, okay then. :) Thank you for explaining your reasoning in more detail. (The edit summary really isn't long enough for this kind of thing.) And you're right, I *was* misled by the corporate page in question. Which is why I posted this in the first place. You're on the ball. --DragonHawk 03:15, 24 July 2006 (UTC)
Apple has ODBC on OS X, maybe someone could add some info on their implimentation? is it just a graphical frontend to UNIX ODBC? 18.104.22.168 07:19, 30 August 2007 (UTC)
Re the original idea. The introduction to this article says CLI was developed from ODBC. However the CLI/ODBC section says the opposite. NetworkComputing () also states that "Microsoft created ODBC in 1992 by extending the CLI (call level interface) from SAG (SQL Access Group, now part of X/Open). It gained acceptance over Borland's Integrated Database Application Programming Interface (IDAPI). The ANSI and ISO (International Standards Organization) adopted an updated version of that CLI as part of the SQL-92 standard, and ODBC version 3.0 aligns with that standard. " In which case clearly ODBC developed from CLI (which then adopted the new ideas). 22.214.171.124 (talk) 12:39, 12 July 2014 (UTC)
This article uses language that is too convoluted and assumes too much knowledge.
The only reader who could get any benefit out of this is a reader who is already very familiar with ODBC concepts and terminology...which essentially makes it useless. —Preceding unsigned comment added by 126.96.36.199 (talk) 12:24, 12 September 2008 (UTC)
"This article uses language that is too convoluted and assumes too much knowledge." This article and the language used within I found useful. I arrived here looking for information, discussion. Thanks to all the professionals here for their efforts to educate. David Blair —Preceding unsigned comment added by 188.8.131.52 (talk) 11:21, 9 March 2009 (UTC)
I provided an overview and greatly expanded the lead paragraph. I'll leave it to the community as to whether to remove the tag, but at least readers who have never heard of ODBC can get an idea of what it actually does now. The article still needs massive cleanup.Quuxa (talk) 13:45, 11 June 2010 (UTC)
XML & ODBC
The article states "ODBC provides the standard of ubiquitous data access because hundreds of ODBC drivers exist for a large variety of data sources. ODBC operates with a variety of operating systems and drivers exist for non-relational data such as spreadsheets, text and XML files.
I was looking for a odbc driver for XML and could not find one. In the reference section, there was not any at either of the 2 odbc driver sources listed. Is this article correct? Does there exist a ODBL driver I can use to connect to XML just like I can connect to a excel spreadsheet? —Preceding unsigned comment added by 184.108.40.206 (talk) 19:36, 25 September 2008 (UTC)
The latest version is said to be 1997. Is this still actively used? I see that it ships with Linux as of 2006, but is this just for backwards compatibility? What alternatives (if any) are there? Mcswell (talk) 03:12, 3 December 2009 (UTC)
Deleted "implementation" section
I rewrote quite a bit of this article because, as a driver implementor, I couldn't refer users to the Wikipedia for good background information on ODBC. The article rambled, used terminology loosely, and made incorrect assertions. (E.g. I've never heard of a non-relational ODBC driver in 25 years of living with ODBC.) I added a section on driver managers, a source of confusion to application programmers and system administrators.
I deleted the list of implementations. It was incomplete and IMO incoherent. The effort to list ODBC drivers is pointless, and the description of the driver managers (and their antecedents) is better left to their respective pages.
I don't know how to remove the "lead section may not adequately summarize" warning. I would if I could. I seriously don't know what else could be stated in the lead -- shouldn't that be lede? -- without getting overly technical.
"a system that allowed calls into IBM DB2 would look entirely different than one that called into their own SQL/DS." Huh, how different? And, entirely so? I doubt this. --Jerome Potts (talk) 01:20, 3 February 2013 (UTC)