Random idea for design collaboration tool

This might be nothing, it might be something. I’ve been getting rather fed up with mailing lists as the primary tool for organizing the Fedora release artwork process lately. Just for fun tonight I did a mockup of one page of the kind of web application I’d like to use to manage the process instead.

Inkscape source – please feel free to download and work it.
Maybe this goes nowhere (except to prove I really ought to find something better to do with my Friday nights…. 🙂 )

34 Comments

    • Hey that looks close to what we need. It'd be easy to hack that CSS but, I'm not sure it has some of the essential functionality we need (like a file bin for source artwork and assets) – what do you think? Do you have any experience with it?

      • …and automatic thumbnail generation for a lot of file types. For this to work it has to be dead easy to submit a design, for example the "inspirations" sidebar looks nice, but I surely won't spend more than 1 minute building it (links, thumbnails, descriptions).

  1. jegHegy

    Excellent idea! Would love to see something like this come to life. Maybe even a global one for all distros that wish to participate, sort of a mashup of your idea, Brainstorm and Gnome-Looks maybe?

  2. I like version made by "George the Dog" most, great wallpaper. 🙂
    Sorry for offtopic, but I have a question about Plymouth theme in F13 – is Plymouth supposed to have different theme each Fedora's version, maybe complement to default wallpaper, or "Charge" plugin is going to be default also in F13?

    • Good question… let me share a bit of personal experience: a week ago at FOSDEM when I meet face to face with people from the large Fedora community I received complaints about the Plymouth theme in F12, which is considered too plain and boring, a big step back after Solar.
      My reply then and now is the same: the Plymouth theme is out of the control of the Design Team, is is at the discretion of the default spin maintainers (a couple of people from the Red Hat Desktop Team).

  3. I recently started thinking in a similar direction for Ubuntu artwork.
    Mailing lists are not good for dealing with files. Having to create and edit wiki pages is a hurdle and if you have nobody constantly looking after it, it becomes a mess.
    A specialized piece of infrastructure could have:
    * templates per category, so you can define what a "page" for a wallpaper or a GTK theme should contain.
    * automatic thumbnail generation.
    * automatic previews with panels and icons overlayed on wallpaper proposals.
    * WYSIWYG editing.
    * mandatory licensing info per submission
    * browse per category/release-target/time-line/author
    * versioning.
    * nested comments plus indication of which version a comment refers to.
    * annotation/adding_marks/drawing_on_top right in the browser.
    Such a more structured environment could be used to encourage a proper design process, starting with clear goals.
    It would be wasteful to implement all this for a single distribution. Hosting content for several distributions could allow to make comparisons quickly and might lead to some cross-talk.
    I'm currently playing with the Google App Engine, which might be a great platform for something like this, avoiding having to care about a server. Sadly I'm a slow coder and if you wanted to realize everything I listed, it wouldn't be a one-man project, anyway.

    • "It would be wasteful to implement all this for a single distribution."
      Ain't that the truth for everything? But when has Fedora ever been guilty of this?
      I certainly would never intend to implement such a thing for Fedora only – I would want it to be FLOSS from the core.
      Anyway, those are great ideas, a lot of them I had in mind. What does 'drawing_on_top right' refer to, is that a reference to something specialized? (it looks like a function name or something?)
      If you crack open the SVG I had an idea for a panel that would pop up if you hover and click… a little similar to the flickr toolbar that pops up above images, to let you try different things.
      Is Google App Engine FLOSS?

      • I did not intend to imply any closedness or not-invented-here syndrom. Actually I started to imagine an Ubuntu branded thing before I realized that it could go one level beyond.
        With drawing on top I mean being able to take an image and scribble arrows and stuff on it. Can be very valuable in criticizing artwork and I figure it would be done more if it was easy and fast.
        There's an open-source development SDK for the App Engine, you can run instances locally on a dev-server, but the real thing, especially the datastore, is proprietary.
        There's an open source implementation in the works, which should be able to run on Eucalyptus. Happens to be sponsored by Google! Don't know how far it evolved so far: http://appscale.cs.ucsb.edu/

        • Oh I love the idea of being able to scribble notes on top. I wonder could you do it so you can show all scribbles, or only show scribbles from a certain person – or attribute scribbles to a particular comment? I'm not sure how to work that out yet. Maybe it wouldn't be a problem – too many scribbles on the same one.
          If the datastore for the Google thing isn't FOSS I wouldn't want anything to do with it. I don't know what Eucalyptus is. I'd ideally want this to be built using TurboGears/Moksha.

          • http://open.eucalyptus.com/ is an open-source cloud implementation.
            If there should happen to be a need for other custom solutions, it would be awesome to a have a FOSS cloud where people can run applications. With an SDK that allows local testing.
            I would normally avoid closed solutions, it's just that Google offers hosting at almost no cost without any server management overhead. That makes it very interesting for me, but I guess you have much less of a problem to get some custom hosting 🙂
            Nice to see that you would chose a Python based solution. Pylons seems to be pretty nice and solid, conceptually. I don't actually know what Turbogears offers on top of it. Moksha is new for me.
            When I briefly looked at Django (mentioned in another comment), I was disgusted by its template language. I understand it's meant to be safe for use by "designers" who can't be trusted from the POV of the developers of a site. In other scenarios, Mako is much nicer, I think.
            I'm just a designer who can code a little bit, at least in Python, not much of a developer, actually, so I better shut up now 🙂

          • I would really hate to implement something so that it only ran on eucalyptus though. Is the API for it specific to eucalyptus or is it portable?
            I just have no tolerance for closed solutions whatsoever – I realize I'm pretty stubborn (but not alone) in this so I'd prefer to keep any proprietary Google stuff far, far away.
            Turbogears & Moksha offer a lot of convenience on top of Python – they integrate all the various pieces (sqlalchemy, pylons, mako, etc) together into one nice system and Moksha gives you cool stuff like live widgets and message bus support.
            Django has a nice pink pony but I'm not sure what else appeals about it 🙂
            I'm in the same boat as you though, I'm a designer who can code a little bit (and hangs around people who are hard core coders) 🙂 No need to shut up I think you've got some great ideas. Just read everything I say with a grain of, 'she's an insanely stubborn 100% FLOSS hippie' 🙂

          • Eucalyptus is compatible with Amazon AWS (EC2, S3, EBS), but that counts for nothing here, obviously 😉
            I have no clue about other open-source cloud stuff.
            This potential project alone is not enough to warrant setting something up and one would have to see what a party doing so would choose.

  4. John McDonough

    Problem is the web isn't push technology. It is way too easy to let it slide and loose track of what's going on. Something akin to Yahoo groups where files are (sort of) integrated with the mailing list might be the ticket. Not that Yahoo groups is the answer, it's just that if you want to replace a mailing list, you need some push component.
    But with you and Thorsten thinking in the same general neighborhood, perhaps some cross-distro collaboration is in order.

    • Hey John, I agree 100%. Uploading stuff manually and having to be on top of stuff is annoying. 🙂 Although a lot of the ideas here I stole from DeviantArt, and artists don't seem to struggle with it, so maybe it would work. I do love automated file sniffing like Yahoo groups does though.

  5. Anders Feder

    I like the look and feel of the page as much as I like the concept itself. It really makes you *want* to contribute.
    I hope that one day, we'll see a whole underwood of tools like this within the open source community, each complementing each other, together making development smooth and fun for everyone.

  6. Cool! a 4chan for designers! it should be fairly easy to implement with a web framework like django; uploading, thumbanils, version control, annotate and even draw on top of an image won't be a problem.
    So… when we start?
    AAAAAAnd… google app engine python interface is based ond django, but sadly isn't floss, and offers less features than django; also, it's google, not something you can install on a server you choose or migrate if google decide to change their terms of use.

  7. Mairin,
    I really think this would be great. While I'm not a designer, I'd love to help in some way with the coding. I'm not great with Python, unfortunately, (I'm a PHP programmer) but wouldn't mind learning if the consensus is to use that.
    Is there another area for me to join in the discussion or has everything been right here in the comments of this blog post so far?

    • Hi Yannick! That would be awesome! right now this blog post is really all the discussion. I think I need to do some more mockups before we have a better idea of what we would need coding-wise. I would really prefer this be a Turbogears app as I can get my head around those and I understand the templating system for that and I know I can get hosting for it. If you're interested in learning TG, have you ever seen the turbogears 20-minute wiki tutorial? It's a good place to start: http://www.turbogears.org/2.0/docs/main/Wiki20/in

  8. Forgotten to mention so far:
    * Interaction/integration with git/bzr(/sv) and code hosting platforms. I'm very unsure about the specifics here.
    * Handling icon sets, including being able to compare how a set of themes implements a specific icon-name (need term to differentiate between concrete icon and the name/role/slot).
    (Ubuntu would also benefit from having some kind of bridge to their wallpaper pool on Flickr.)

  9. Mairin,
    This is Josh with Isotope11. We're a rails development shop, and we've been looking for tasks for a new hire to do pro-bono to get up to speed with our development processes (as we don't like the idea of charging for what is essentially his training time).
    This project is fairly similar to his first project that he's finishing up, and it would be a useful tool for our company to have in its toolbox. We obviously deal with clients going back and forth on design/ux and I like the layout you put together for this tool.
    When he's done with his current task (today? Tomorrow?) we're planning on him beginning to implement this product based on your mockup. The early version would likely not be feature-complete, and the finalized version would likely lack some features you may want…but it will be open source.
    Could I interest you in a dialog of some sort to discuss the software a tad beyond the mockup as given? You clearly have a firm vision of what such a tool would look like, and it'd be foolish not to pick your brain for it 🙂 Just feel free to email me to discuss it further. It would run on any platform that can run a rails app, of course. Hope to talk soon,
    -Josh

  10. I love the concept. The biggest problem I have with the convos that go on the mailing list, is following them, and all of the "headache" that goes into seeing the iterations. This layout makes it easy to see where the design came from, and gives it more context for new people and people that may have deleted previous emails.
    I also can volunteer my programming experience I'm "fluent" in php, perl, python and while i don't know turboGears or ruby at this point learning a new language is little more than learning syntax.

  11. Pingback: The one where the designers ask for a pony « Máirín Duffy

  12. Pingback: Another Design Hub mockup « Máirín Duffy

  13. Pingback: Design Hub Update « Máirín Duffy

Leave a Reply

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.