Getting the picture – an interview with the Graphic lab
Vectorisation of raster graphics. One of the most common tasks requested at the Graphics lab.
This week we met up with five wikigraphists—members of of the Wikipedia Graphics Lab. We chatted with JovianEye, Begoon, Fallschirmjäger, Gringer and Orionist.
Unlike a conventional WikiProject, the Lab does not work with articles – only images, grouped into three "workshops": one for illustrations, one for photographs and one for maps. When the project started in 2006, requests were all grouped in one area, as the Graphics lab. Since then, the three separate workshops have been created, focusing people with relevant skills on the specific area they can work on rather than having to look through all requests. However, this sometimes means keeping watch on three different pages, each with a lower level of activity, as well as the sister project on Wikimedia Commons.
The project is in need of more people requesting improvements to photographs. However, there is a backlog at SVG creation requests, and a shortage of skilled editors in this field. With only a few regulars, notably veteran requester Chris, the Lab is only scratching the surface of the 7 million files on Commons alone. The addition of the 'Top 4' feature at the start of this year, a set of images requiring vectorisation at the top of the page that are changed every 48 hours, has helped in encouraging progress and getting through more files.
The project would also like to encourage beginners to become Wikigraphists by learning to use simple freeware applications, GIMP and Inkscape. Simple tasks such as cropping can be learned by just about anyone. They can try their hand at removing borders or captions of some of 80,000+ images donated to Wikimedia Commons by the Bundesarchiv (German Federal Archives).
Removal of objects and color improvement.
Whitening or cleaning of the background, another common task.
An example of an image created with Inkscape
, one of the free programs used by many wikigraphists.
are programs often used for the retouching of images and creating raster graphics
. These programs were used to create this image.
Jovianeye made his first request at the illustration workshop in January this year, and describes himself as having a ‘basic skill level’. He has since also made contributions to the Photography workshop. For him, the main problem that the project has faced was when Fred the Oyster, a professional graphist who volunteered almost every day, was blocked indefinitely. Fred's contributions to Wikimedia Commons can be found here. Jovianeye was influenced to join the project because the requests at the Graphic lab did not need a long-term commitment and because they could “be handled by anybody with a beginner skill level.” Without a background in graphics, it's “merely an interest”; he grins and says that “even the word hobby is too strong!” Since joining the Graphic lab he has been inspired to learn the basics of open source vector graphics editor Inkscape.
Begoon became active about three months ago. He does a large number of SVG conversions for WikiProject Scouting, which, he says, can often be interesting because they have a huge number of images, in various conditions, from all around the world and different time periods. He enjoys learning about places and times as an offshoot from taking on these requests. Begoon is a programmer and web designer, so deals with graphics on a daily basis. He is more skilled in the programming aspect [of web designing] but being independent, has to do graphics work. The requests in the lab are often challenges that help him to develop new skills and techniques, while learning how others approach a task. He thinks it is a “good way to hone his skills”. The project suits his free time pattern, too, "because you can take on as many, or as few, requests as time permits”.
Fallschirmjäger started work in the project back in April 2008 (see the archives), although editing images is his main area of contribution since he joined in 2006. He works mainly in the Photography section of the lab, but often helps out around the Illustration and Map workshops. He has worked on many files in both Wikipedia and Wikimedia Commons. For him, the main problem is a lack of skilled editors, particularly in the Map Workshop. Although Fallschirmjäger did try to write articles when he first joined, he soon found that it wasn’t really for him, and turned instead to image editing, although he still contributes occasionally in writing and minor text-related edits. He has used Photoshop and other multimedia software for some time, and has also recently completed a degree relating to the field.
Gringer discovered the Graphic lab about a month or so before February 2009, although he admits he had been patching up a few images since January 2007, for instance File:Real grounds 1000.svg. He likes the split of the Graphic lab into different workshops, even though this is occasionally ignored by requesters who notice that particular categories have better turnaround times. More requesters are needed, he says; as each requester has a particular area of interest, and he feels that the project is not seeing a representative sample of graphics that need fixing. Gringer found that cleaning up images was something he could usually do quickly. He says he's very much a “hobby wikigraphist”, but tries to create his own images for presentations he makes. He often uses sources such as Wikimedia Commons and openclipart which give him practice with Inkscape.
Orionist took his first request in May. He first worked on photographs, but is now concentrating on illustrations, because there’s more need for his skills and the photography workshop is very well covered by editors. Wikipedia and Wikimedia Commons have been an invaluable resource for his work, where he could get images that were not available on any stock photography website. He is still an active editor on Wikipedia, but finds that images require much less time than article editing, with his busy daily schedule.
Individuals' goals for the project
Fallschirmjäger wants to expand into the other two workshops. Currently, he works with bitmap images—his main area of expertise—but he would also like to help out with vector images. He thinks that a set of advice or tutorials would be a worthwhile undertaking to try to get more people involved in the project. He thinks other projects aren’t always too concerned by images in articles within their scope, yet they can add a lot of value to the text. He wants to see more activity in the project and more involvement with other editors and projects. One of Gringer’s goals is to try to find some free, up-to-date database of country/administrative region borders and an accompanying script, so people can generate SVG maps of any region in a few different projections. He has noticed there are a lot of variations in sources used for maps that in some cases have automatically dried up. When Orionist completes a request, he takes a look at its article and Wikimedia Commons category to see if there are any related images that need to be edited in the same way. Thus, some requests have evolved into personal projects.
What help is needed?
Jovianeye feels the Map Workshop needs more volunteers, to prevent requests from becoming stale. Both the Illustration and Photography workshops are doing well, he feels, however the participation of more Vector graphists is needed at times. The project does have highly skilled professional graphists, but they are not always on hand to deal with requests. Begoon tries to extend his involvement in the Graphic lab from Vector images to photographs as well, since he wants to gain expertise in a wide range of image editing skills. He agrees with Jovianeye that the Map Workshop needs more help and intends to explore the skill-set involved in Map editing soon. Fallschirmjäger also thinks the Map Workshop needs assistance. More work is done in the other two workshops because even the hardest of requests generally take no longer than a few hours. Working on maps requires a certain amount of research to ensure that the map’s labels and boundaries are correctly detailed. This is largely why there is a backlog of requests and a lack of willing volunteers. Gringer notes that it would be nice if librsvg had better SVG support, which is an issue with Wikimedia Commons, not the Graphic lab. Then volunteers wouldn’t have to spend so much time debugging SVG files so they look nice on Wikipedia. Fonts, he says, are “a frequent problem—Arial is pretty much the only one that works.” He says advocacy could help the project, both for more requests and for more wikigraphists. Orionist says more people need to be made aware of the lab’s existence. He points out that he has been on Wikipedia for four years but only recently found out about the lab. He suggests featuring it more prominently in help pages, running an ad on Wikipedia, or posting messages on talk pages of articles with retouched images ("This image has been retouched/made in the graphics lab!" or a similar phrase).
Each of the guys has altered many images while members of the project: Jovianeye has altered at least 50 images in the Illustration Workshop and at least 10 at the Photography Workshop. However, just like an edit count, this number can be meaningless. Certain requests require a great deal of effort and can easily be equivalent to doing 10 simple requests! Begoon's image edit count is somewhere between 150 and 200. Fallschirmjäger says he has made edits to hundreds, if not thousands of files. He also adds that since requests range from a simple crop or rotation to full-blown image manipulation and creation, the time spent to get each right can vary significantly.
Tips to requesters
Gringer says: “Requests are fulfilled in my own spare time, and I don't get paid for it. If I think something will take too long to do, then I won't do it, so anything you can do to make that process easier for me (e.g. finding good source images) can increase the chance of a successful conversion.”
Next week we'll inventory Wikipedia's rolling stock. Until then, board the next coach headed toward the archive.
This image was created using Hugin which is used for the stitching of panoramic images together into one big image, correction of perspective and various other things.
An image created by Blender which is used for the creation and editing of 3D images.