Collaborative tools for designers: part 4

TL;DR: Tools I will not review and why.

Three weeks ago, I started a series of publications regarding tools that encourage collaborative practices for designers and that use some form of versioning at their core. You can read the previous chapters and especially the first one to understand what I’m looking for (links to part 2 and part 3). In this fourth episode, I’m just going to list tools I will not review thoroughly but that I felt I should mention because they exist (or existed) for certain reasons. If you have experience with some of them, I’d certainly be happy to hear about it.


Used by blockbuster media companies (such as Rockstar Games or Disney, for example), it’s a complete project, team, assets, source control management solution aimed at very large studios with hundreds or thousands employees and its apparently great at handling large (sets of) media files (at least, that’s how it was once presented to me). The licensing fee are relative to that description. It’s worth noting that the latest presentation of Perforce Helix mentions the integration of Git for the developers in the team. Maybe worth watching is this video showing how to handle their image diff tool.


Although, like Pixelapse, I had them on my watch list, I never took the time to dig deep into their product. And unfortunately, as I understand it from the description on their website, the company went bankrupt 6 months ago. Too bad, when you see Kelly Sutton present their product just 2 years ago, it looked very interesting. Their “wormhole” tool (at 6:00) is really a nice approach to get “what changed and when” in an image.


I hesitated for a while. Should I review this tool or not. They call themselves “The world’s leading prototyping, collaboration & workflow platform“, and looking at their client portfolio, they surely can pretend to this, I guess. But they focus mainly on screen based media and interactivity, or so it seems. and their level of version control is no better than Dropbox or similar. So the designers targeted here are “application” designers, UX/UI designers,… and I’m looking for a a less opinionated tool. Anyway. If you want to get a quick glance at what this tool does, here’s a quick introductory video. Certainly a great tool to review designs, manage projects & teams and test user experience. Not what I’m looking for atm.

Visual Culture

Visual Culture is a server software developed by Open Source Publishing to visualize git repositories. Since their design practice involves doing things open and in the open, their website runs on it and allows the visitor to browse into their latest projects and how those are made. Their intention was to turn it into a general tool for designers and they opened a crowdfunding campaign to achieve that goal, but this did not received enough support. I did try to install it and see how I could use it on my own projects. With the help from Colm, Stéphanie and Eric, I passed through a couple of bugs and misunderstandings on my part. And I could see a recap’ page of what I had in my folders. Unfortunately, that’s how far I managed to go. I lack some Django knowledge to adapt the necessary bits so it would behave nicely with my projects as, at this stage, Visual Culture is still very closely knitted to (how) OSP(‘s) work.

visual culture on my repos

Come back next week for another chapter.

Collaborative tools for designers – Part 3 : Pixelapse

Reviewing tools for collaborative practices and version control for designers, this is episode III. And to my surprise, I’ve discovered Pixelapse, an online tool that promises to do just that.

Pixelapse Homepage

It’s been a while since I had found about Pixelapse, but never took the chance to study it deeper until last week. Mainly because they require to install some software that only works for Mac and Windows. And since I use Linux, well…

So I borrowed a Mac without much luck either. Pixelapse software requires Mac OSX 10.7 and higher. The Mac was running on 10.6. I know designers have usually high end machines with latest software installed. But no Linux support and only a recent OSX, that’s 2 wall hits for me before even starting. But anyway, solved this by installing a Ms Windows running in a Virtual Machine under Linux because I really wanted to see how their tool worked. A pretty big setup just to sync files and not practical in the long term, but for the love of testing, it was fun enough.

Basically, what you are installing is a synchronizing software. Just like Dropbox, it creates a special folder on your machine, where you will keep your files synchronized with their servers and with the other computers you share that folder with. The other computers could be yours or from the designers that you want to work with. Apart from that, all the features proposed by Pixelapse run in the browser. Well, not every browser. Seems they only fully support Chrome. Internet Explorer is a no go and I don’t blame them for that (I tested IE11 only). Firefox seems to work ok, except for the “image inspector”, and that’s a pretty big feature being broken for no apparent reason.

visual comparison

About the features, they have really though it through. You can see their service as being between a version control, a project management and a reviewing tool. You can define roles for each projects, invite “clients” to review your graphic files and comment on them, or have collaborators with {full|limited} permissions. They support a wide variety of file formats for preview (mostly Adobe stuff), but also open file formats and even .xcf (Gimp files), although the latter did not render text properly. Version control is basic (no concurrent versions and they don’t allow custom naming your versions, so you’re stuck with version1,2,3,…) but at least they allow you to rename and move files around without losing their history and you can flag “milestones” in your project’s history. They also allow you to visually compare files or file versions in different ways. Commenting and reviewing design seems pretty well made, Although I could not test it on an actual project or with a team as I don’t know anyone who’s using this tool.

So my overall feeling is very positive, but something is bothering me.  Apart from the fact that it’s not open source, not working on all systems and that you have to trust them with your files, why did I not hear of them more? If I had to do a comparison, I’d say this aims to be the Github for graphic designers, but why don’t I see more of us use it? As a matter of fact, I will definitely find more graphic design projects on Github than on Pixelapse. Why is that? They do allow public projects for free accounts. They encourage in their communication to do open source design projects. They feature some on their homepage. But after going through what was available publicly, I did not see much collaboration going on (Googling through the website, I found one public project with 8 participants.  But can’t figure out if there was any collaboration at all. And then this UFO, which seems to put to good use the features proposed, but I’m wondering if this was really meant to be public). I know it’s a young service (2 yo), but Internet is pretty quick to embrace a pretty good idea when it sees one. Where is it not catching up?

Pixel diff between two versions

I don’t have a definitive answer, but my feeling is that the interactivity is a little oversold and that it unfortunately still enforces the digital divide between coders and designers.

This tool is for graphic designers only, coders will not feel at home. Yes they support the viewing of some code files (HTML, CSS, JS), but without line numbering and a basic text diff tool, this is useless, even scary. Also, why not support other languages like, Ruby, Python or Bash scripting? Where is the problem with that? There are some good ideas : like having a file and a cover image at the root of your project to automatically create a nice presentation page for your public projects. But then, without a proper way to see what changed in your markdown syntax, this becomes a little awkward to use.

Designers of pixelapse

The navigation has problems also. Exploring public projects is a pain. There is no search bar. You can add tags to your project’s description, but it’s like they forgot to implement it on the backend. No way to browse based on project type or keywords. No way to see the latest activity on the site resulting in no sense of community. Seeing the public projects they feature, it feels more like Behance − or any portfolio you can name − than an “Open Design Movement”, as they want to call it.

The interface is also a drag sometimes. It’s using animations here and there and this makes it all a bit clumsy. I want my tool to be fast and to the point. Animating between a thumbnail view and a list view of my files is not something I find important or informative. I wish also that space would be used more efficiently. The whole window is used when previewing files, comparing and reviewing them. And this is a good thing. But why do I feel that so much space is lost when I’m navigating folders and projects. Why not have the full screen used also for that too? I also found myself clicking a lot − where there was no button − expecting something to happen. At other times, I was completely lost and had to go back to the main page to start again.

Too much white space?

Maybe it also tries to embrace to much. Working with clients is not the same as working within a “private” team and not close to working with strangers on open source projects. Keeping designers distant from coders is a also counter productive approach in my opinion.

The good news (or the bad news, depends on the point of view, I guess), is that Pixelapse has just been acquired by Dropbox. So it could mean that Dropbox will improve or that Pixelapse will support more operating systems. I’m guessing Dropbox is interested in the visual reviewing tools since they’ve just started introducing a commenting system for their files. As for the future of Pixelapse, Here’s what they have to say about it:

Will Pixelapse continue to operate?

Pixelapse as a standalone product will continue to be available and supported for the next year as we work towards bringing the same kinds of collaboration and workflow experiences that you’re used to in Pixelapse over to the core Dropbox product.

That just sounds to me as “good bye and thanks for all the fish”. I don’t expect Pixelapse being updated for the next year and have not seen a post on their blog or twitter since the Dropbox announcement (and that was 6 months ago).

There is a ton of good ideas in Pixelapse. But it needs polishing and a different approach  regarding their community if they really want to be the repository platform of the “open design movement”. So for me, I’ll just pass and move on.

Collaborative tools for designers – Part 2 : Dropbox

rip dropbox

Last week, I promised I would be exploring some of the tools promoted as helping designers in their revision control and collaborative process and that I would do this by comparing them to a list of benefits I extracted from my use of Git. So in this post, I’ll be reviewing the (in-?) famous Dropbox.

As a foreword, I have to tell you I don’t use their service. I was forced to have an account, when some friends and coworkers wanted to share some files with me, but I never liked the idea of having my draft files on a computer I did not own. I use some third party services to share publicly some projects, as you will find on Github, for example. But by default giving to a stranger a work in progress, some first sketches or an unpublished project, is not something I find comfortable. How good or easy the service could be.

So as a start, this rules out the use of Dropbox if you work under an NDA, have company secrets or just care about the keeping your working process to yourself or your collaborators. But let’s analyze what they offer and see how that helps designers or not.

First of, it’s an easy set up. After a necessary software install, sharing a project between collaborators is as easy as moving around a folder in a file explorer. It’s also “operating system agnostic” as it will work on Mac, Windows or Linux and even Android or iOS. And if that’s not enough, there is always the web interface to rely on. Actually, the web interface is where most features are built in. (I’m testing this from a Linux computer, but I guess it behaves similarly on other systems).

On the “workstation side”, there is not much more than moving around files, renaming them, and modifying them with your software of choice. It will then synchronize those changes automatically with all the computers you share that folder with.

On the “web side”, Dropbox offers you some form of “versioning”. Everytime you save your file, Dropbox records a change. You can go back in the history of those changes in the web interface and see who made them, from which machine. But unless you know what changes where made and when, there is not much more info given.

dropbox file history
See for yourself, that list of versions is not that useful. If I did a mistake and saved over my file, I could use this to immediately revert to a previous version. But come back to that a week or a month later and who knows what’s hiding behind “version2”. If the file type is supported by Dropbox, I can see a preview by clicking on the version name in the browser. If not, I’m presented with a download button and have to open it on my computer. If somehow we could add a comment or a note with each version, that would improve this a little.

Also, there is no way to compare different versions (even if it’s a plain text file) except by viewing each independently − opening multiple browser windows − and figuring out for yourself what might have changed. Forget also about having different concurrent version of the same file. You’ll have to create different files for that and you’re back to your weird naming scheme. And if you rename your file or move it to a subfolder, your going to be losing all its history. Really? How useful is that?

So this revision system is more like a notification system of who worked on what and when, than a reliable way to store your changes as this can be wiped out by just renaming a folder. It does not help you get organized either. And I found the links in the “events” page confusing and practically useless, especially if you have a lot of files and subfolders in your main shared folder. (see screenshot below, all these files are located in different subfolders in a master folder called “libre-objet”).

dorpbox events
(For some reason, I notifies me of changes on some these files, but when I’m clicking to view the history of it, it only says “version1”. Was the file renamed at some point, or the containing folder, or because those modifications were done before the folder was shared with me,… I have no idea. But how can I trust this?)

I did appreciate though that I could get a 1:1 preview of some files like SVG, Tiff, Jpeg, png, ai or psd (I did not review all the possible files formats). That 1:1 preview also features a commenting system, where you can tag collaborators. But I did not see any tool to annotate a design for example. Maybe that’s coming… who knows.


It also shows a little thumbnail for some of those file types. That’s handy. But this is also something I get in my file explorer on my computer, so where is the added value here? Why not also have those thumbnails in the history page next to each version, or in the event page also? That would be a little more helpful.

As a conclusion, Dropbox is maybe, for most, an easy way to get a shared folder on many different machines but it can not be trusted. Neither on the conservation of files, nor on keeping a history of changes. If you don’t care about “versioning”, you’ll be better of with alternatives, such as BtSync (closed source) or SyncThing (open source) and all the other self-hosted “cloud” application suites.

I know I’m killing down here a great service for some of you. And Dropbox never pretended to be the git (or github) for designers, even if they seem to be heading in that direction somehow. But their versioning system is a joke. And I’m looking at all of this from a very particular point of view. Please, don’t hesitate to share your designer’s experience using Dropbox in the comments below, I’ll be happy to read it and it might help me understand parts I could have missed. Again, I barely use the service. And if you enjoyed this post, come back soon for part 3.

(Cover photo by Steven Caddy released under Creative Commons by-nc)

Collaborative tools for designers – Part 1

Soyuz Cockpit

Since I’ve discovered and used distributed revision control in my day to day practice, I’ve been amazed by how much it has helped me getting more organized, stop worrying about loosing files and also encouraged me to experiment more and thus try more options in my work. And I’m not even talking about how it eased the collaboration process with other designers and coders.

But one of the major drawback is to convince other people to use it. Convincing someone who codes already, or who is not afraid of typing commands in a console, was easy. But talking just pure GUI fanatics to jump into it, is a bit harder. And for a good reason. The tools needed to manipulate the usual revision control systems are mainly designed for coders. Yes, there are some GUIs (I don’t use them), but they look like a spreadsheet or a Soyuz cockpit. And that barely helps make sense when you are looking at the type of projects that graphic designers have to cope with.

So, I’ve been trying to put my head around this problem and come up with solutions. I had a first try at it with a project called “Design with Git” that I had the chance to develop at Medialab Prado, Madrid in 2013. But this was just a beginning, a short-time experiment, and although I’ve spent energy applying to other residency programs to continue on building it, I’ve yet to find a place that would allow me to dedicate some time improving this concept.

So, since this subject is not dead for me and for a lot of others, I’ve decided to start a series of blog posts, in which I would clear my head out and put into words and graphics my thoughts around this subject. And as a start, in the coming posts, I’m planning to review some available tools for “revision control” and collaboration between designers.

A word of warning here, maybe. I’m biased by my experiences, of course. And so far, the only version control system I’ve used extensively is Git. When I started being interested in these kind of tools, I got my hands on CVS, but then Git (and Github) quickly became the trending thing and I drank the Kool-Aid. Also, CVS needed a server, and that was too much to set up for me just to manage my own projects. I’m not a Git-guru either. I use it, almost daily, for almost all the things I produce. But I’ve not gone beyond the basics of cloning, branching, diffing, pushing and merging. When I have to do more complex tasks like going back to an older version of my files or completely erase a file from history, Stackoverflow is my friend.

So here is a list of what I find useful when using Git with my projects:

  • It allows me to view a history of changes, which files have changed and a written description about it (status / log)
  • It allows me to easily make different versions of the same project and to quickly switch between those versions (branch)
  • It allows me to work with an undefinite number of people (clone / fetch)
  • It returns precise and valuable information between the changes or different version of the project (diff)
  • It allows me to integrate the changes made by another person into my own project (merge)
  • It does not require a complex setup to start using it (init)
  • I do keep my folders tidy as I don’t have to invent complex naming formats for my files (No more Front-cover-v4a-TOPRINT2-validated.pdf)
  • It allows me to view the contributions of each participant on the project (log)
  • I’m not worried about loosing a file, erasing or writing over my previous work.
  • It eases the process of working with developers because we are sharing the same tool to keep track of the project’s advancement.

This list, I think, sums up for me what are the main benefits of using a distributed version system such as Git and these are the benefits graphic designers should get from using it also. The catch is this works well when you’re dealing with code or text files. But some points here are a little or very problematic when it comes to dealing with design files. Hopefully, there is some solutions or things that can be adapted. And I believe this is possible while keeping Git at the core of the process.

But before going further into this development, I’m planing to explore some of the tools promoted as helping designers in their revision control and collaborative process and I’ll do so using this list as a frame or a grid of evaluation.

So see you next time. (go to Part 2)

(Cover image: Soyuz cockpit by Christopher Michel under Creative Commons Attribution 2.0)

Constant V (Venster/Vitrine)

Ami Bruxellois, ne loupe pas l’occasion ce vendredi jeudi (18/06/2015) d’aller voir la vitrine de Constant. Tu y seras bien reçu et tu pourras même y boire un verre. De plus, c’est juste à côté du Verschueren, si jamais ta soif était trop grande.

Constant V - uHbench

Je t’invite à y aller, pour découvrir Constant et pour y pêcher un petit manuel afin de construire toi-même un banc public à partir d’une palette. Tu pourras même t’essayer à la fabrication puisque du matériel sera prévu sur place.

L’atelier sera chapeauté par la main experte de Barbara Janssens.

Constant V (Venster/Vitrine) est une série d’exposition dans la vitrine de Constant. Les objets, les dessins, les vidéos, les installations et les expériences exposés sont sous licence libres ou ils ont été créés en utilisant des logiciels libres.

Constant V (Venster/Vitrine) is een reek tentoonstellingen in de vitrine van Constant, De kunstwerken variëren van Open Source objecten tot video, design, en artistieke experimenten met vrije software.

Constant V : uHbench
Expo: 18 – 28 juni/ juin 2015
Perfo + drink: 18 juni /juin 2015

#uHbench map

A map of known places where urban hacker benches have been built.

uHbench map
A workshop and exhibition of the uHbench project has been scheduled in Constant VZW at the end of this week. For that, they’ve asked me to produce 2 posters that will be displayed along the benches built for the occasion.

One of the poster was obviously going to be the instruction sheet. But for the second one, I hesitated for a while, and finally set up for a representation of how far this project has reached since it’s creation in 2012.

Because thanks to the open source nature of this project and the support it received − not only from friends but also from members of diverse communities  all over Europe −, this project has traveled further than what I would ever imagined. So this map is, of course, a little self-flattering, but I do hope it creates a sense of community beyond borders for those who participated and maybe will encourage some to add their mark to the map.

The vector map is here:

For the technically minded, the map was done using the wonderful D3js library. Which I was discovering for the first time but enjoyed a lot (amazed by its possibilities). The code for this project is shared on Github.

− Have you built a uHbench? Want to have your instance on the map?
Add an issue with the name(s) of the builder(s), location and date or clone the repo and update the data.tsv or send me a mail with those informations.

Blender shortcuts right in your search engine

Blender Cheat Sheet

Two months ago, I attended a community meetup in NY called Quack & Hack. The point of it was to gather people around coding to improve the DuckDuckGo search engine.

If you don’t know DuckDuckGo (DDG) by now, it’s the “search engine that does not track you”. And it’s been my tool of choice when I want to search something on the web. It has nice features like the !bang mechanism and the instant answers which I find really handy. But most importantly, I feel good not being followed by an all seeing eye like You-Know-Who.

Another great thing about DDG, is that they have a program called DuckDuckHack where they invite coders to submit improvements to the search engine. These, if approved, end up for example as being new instant answers. I, personally like the weather one, the password generator or the Gimp cheat sheet. But they have many many more. It’s impossible to know them all.

So, as a user and fan of Blender, I thought it would be nice to have the crazy number of shortcuts for that open source 3D software directly available in my search engine of choice. It would not only be helpful to me, but maybe also to the larger Blender community. And in the end, it might also have some of those Blenderheads care a bit more about not being tracked when they avidly search for the latest hot features of Blender.

In the end, the process of getting those shortcuts live on DuckDuckGo took longer than I expected. But it’s now live since yesterday. So try it out and search for “Blender cheat sheet” and tell me what you think.

You can take a peek at all the effort it took by looking at the conversation I had with the DuckDuckGo team on Github. What mainly happened is that I first submitted a “Blender Cheat Sheet” using the classic way for submitting instant answers and during that period, they were working on a different one, which would be easier just for submitting “cheat sheets”. So I had to port that to the new system, and adapt the code until their new stuff was stable and ready to ship.

But I’m glad to have gone through all this. It was an interesting experience to work with the DuckDuckGo team, a nice and friendly crowd. It also feels rewarding to know that my little contribution might hit thousands of users. And I’ll be glad to have that feature in my toolbox when I’ll be giving the next Blender workshop here in NY, at the end of June.

Blender Cheat Sheet all

I finished Facebook #orSoIThought

Facebook High Score Animation

On May 27th, I’ve finished The stupidest game on Facebook. I was silently running this little self-amusing game since the 15th of May and then, later on, opened the process publicly. In the end, it took me less than 15 days to fill up to 99 the most famous 3 icons that you find on the top right corner of Facebook… on a desktop browser.

99+ 99 99 So that’s what it looks like. No cheating. No Css rewriting. No photoshopping. And as a proof, I was screen grabbing the page every 2 hours from the mobile version of Facebook. (Actually, I did not set that up for this project specifically. It is a part of my LoveMachine project, which has been running for more than a year now: I take a screenshot of Facebook every time the bot runs, mostly for debugging.)

So today, I looked for the first time at all these screenshots and especially the ones corresponding to this period and to my surprise, I’ve found out that the stupid game does not work at all on the mobile version of Facebook. If the highest score on desktop is clearly 99, I have no idea what the limits are on mobile. (On the top of this page, you can see the gif created from all the screenshots.)

This is of course very disappointing. It makes the game a little useless, or at least, not multi-platform, as games should be these days.

Anyway, I guess that closes the challenge. I have other plans for the monster than trying to reach 999 or whatever the limit is on mobile. If any of you is up for the mobile challenge, please drop me a line. I’d be interested to see your results.

As a happy note, and to close this chapter, I found out that when the score of unread private messages reaches 99+ in the desktop browser, it resets the counter by itself. Which, to me, is obviously a bug. So, as a joke, I posted a status about it and tagged the Facebook Design team in it.

Facebook design response

They replied.

The stupidest game on Facebook

and you don’t need to install any app for it.

Facebook Highscore - almost there

While contemplating the most watched icons on Facebook, it became obvious to me that this was an actual Facebook score. I’ve always believed Facebook is a game. And therefore, it should not be taken seriously. Any other use of Facebook would be foolish.

So these icons, well placed on the top right corner of the window, just like in video games, are supposed to give you information, but they mainly only invite you to click. For those not on Facebook, the left one is for “pending friend requests”, the middle one is for “unread private messages” and the one on the right is for “unread notifications”. That last one behaves more as a garbage collector because there is a lot of different kind of notifications. It can be comments, likes, game notifications, tagging, etc.

The trick with these icons is once you click on them, it resets the counter. It doesn’t mean you have read the notifications or that you replied to all your messages or that you responded to all the friend requests. No. It resets because you clicked on them.

Another thing to note is that the counter for each of these icons stops at 99 or 99+. There’s an end to it. Beyond 99 friend requests, private messages or notifications, is not worth counting. Why? Why not 999, 9999…? I sometimes have my mailbox or my RSS reader saying 157 unread messages. But no, not on Facebook.

Anyway, this led me to start a game. A stupid one I admit. The game of “not clicking” (looking at you reddit). I promised myself not to click on those icons until they reach their maximum. And I dare you to do the same.

So far, the easy part was getting friend requests and notifications up to the top. Now, I’m slowly waiting for the private messages to come in. A little more than 20 to go. I did screw up two weeks ago when reaching 42, I clicked on the chat window to respond to a friend. That also resets the counter… But thanks to friend support and announcing publicly what I was playing, things got back on track.

If you want to play along, just take a screenshot of those notifications icons and post a message with the hastag #FacebookHighscore

Let’s see if Facebook breaks when we all reach: 99+, 99+, 99+.

RaspiVj (alpha release)

Yesterday, I’ve pushed online the code for what I call the RaspiVJ, a minimal VJ application for the Raspberry Pi.

RaspiVJ set up

The idea was to build a portable and lightweight VJ set up that could be versatile enough to adapt to multiple situtations and configurations. So the Raspberry Pi, as a small and cheap computer, seemed a good candidate. It can already play videos up to an HD format and can connect to a screen via HDMI or RCA (for those old televisions).

Inspired by PocketVJ, everything runs from the Raspberry Pi. All is needed is a smartphone, tablet or computer to connect to the web interface of Raspivj and then be able to manipulate the videos and switch between clips.

RaspiVJ screenshot

The difference with PocketVJ so far, is that it behaves more like a VJ application. You can quickly switch between clips, crossfade between two videos and fade to black.

I’ve already been using this application during a live show with the band Left Arm of Buddha. And I think it could be the perfect tool for young bands, performers or musicians that would like to play videos during their shows with a bit of control over the playback, while still being simple to use and easy to set up.

Left Arm of Buddha

Technically, a node.js server sends the interface to the user’s browser and communicates any action from that user to a custom video player (coded with openFrameworks and ofxOMXPlayer).

So all you’ll need is a Wifi dongle for the Raspberry Pi (or a network cable), some cables to connect the Raspberry Pi to a screen or projector and a smart device to access the webpage that acts as the interface.

The code is released on github as open source under an MIT license. Right now, you have to go through a somewhat lengthy process to install the application. But I’m planning to make a plug-and-play image that you will just have to burn on an empty SD card to get running. Just ping me if you’re interested, it would increase my motivation to do so.

Else, if you just like this project or wish their could be more features, don’t hesitate to say hello.