WIPUP.org – aiming small

Today I wanted to talk a bit about the birth and objectives of WIPUP – a subject I haven’t really revealed before. WIPUP, for those who aren’t already familiar with it – is an open-source web application I created which allows people to document, share, track, and critique their works-in-progresses, or in short, WIPs.

The project began quite a while back. I had made relatively significant progress on the ThoughtScore project – my hobby animated film, and I wasn’t content with sharing it on the BlenderArtists forum – it seemed very limited and non-specialised for project documentation. I had also had the VisionBin project – a portfolio-generator webapp – running for a few months. My “finished” renders didn’t fit well there either. VisionBin had been running for some time and it wasn’t doing too well – the concept wasn’t differentiated enough and it didn’t perform its task particularly well either (especially in hindsight). I also hadn’t touched programming for a while and was getting a tad rusty.

That’s right – the time was ripe for a new project.

I evaluated the situation and decided that I needed to make a system dedicated to sharing the in-between. Not the mini-projects and small-time creations which forums, blogs, twitters, deviantarts, etc, were fine for, but also not for massive projects which were kept under wraps until they were unveiled – for everybody to enjoy the finished project but disregard the beautiful, hidden, shunned process behind it. I needed to expose this beautiful process. This was the key behind keeping ThoughtScore alive. Turning the arduous learning process behind an impossibly ambitious project into something to be celebrated. This – yes –  this was WIPUP.

As you can see, WIPUP was a very selfish invention. It was a system for myself. I wasn’t interested in communities or distribution. In fact, the first release on WIPUP wasn’t built on the open-source Kohana framework, but instead on a company-tied solution called CodeIgniter, and that WIPUP release was closed-source.

It was only later when I was frustrated with some of the slowly developed aspects of the CI framework did I have a discussion with the folks in the Kohana channel, and WIPUP was half-built did I begin the Eadrax project. The Eadrax project was the open-source rewrite of the then alpha-quality WIPUP under Kohana. That was the time I decided to share this system – and guest WIPs and user accounts were added to the system

During the development of Eadrax, I was exposed to similar projects such as Dribbble, the *bins (temporary WIP hosting), and various others I can’t recall right now. They still didn’t suit me – they lacked flexibility and organisation. For flexibility – most were very oriented towards a very specific format – an image snapshot, a sound upload, etc. None offered the flexibility to have an update to be a simple as a Tweet to the complexity of embedded video and multiple image attachments. I worked on a huge variety of projects and just supporting one but excluding others wasn’t good enough for me. Similarly, for organisation, none seemed to offer any form of proper project categorisation. I needed a way to separate out my work into projects – view my progress as a whole and split within projects. That was how projects in WIPUP were introduced (they were also taken from VisionBin)

Finally – I needed my data to be free. I didn’t want my careful documentation of my personal projects to be lost to a third-party, forever bound within the constraints of their system. I needed to be able to retrieve it however I wanted and format it as I liked. None of those systems were open-source or offered any form of security. WIPUP then turned open-source, and implemented the Open Collaboration Services API, and is now looking towards project export capabilities.

Much more interesting than these dissatisfactions was the realisation that my needs were – perhaps sadly – a rarity. Count the number of people you know who has a hobby where they create stuff which can be shared. Now within that group of people, count the number whose hobbies have sufficiently long-term work-in-progress periods such that it makes sense to document the process. Already we have a very small number of people, if any. Then, out of the remaining few, pick out those who have multiple concurrent projects of varying nature and characteristics. Almost nobody? Perhaps one or two? Finally, single out those who actually want to or can share this process. That’s the killer. Most can’t. Most don’t want to. Perhaps they’re restricted by a group project or by a company. Perhaps it’s such an amazing project they believe it should be kept completely secret. Perhaps they don’t see the point. Perhaps they don’t have time, or are too focused on the finished product.

I then followed through this realisation by testing out if there really were people in everyday life that thought like me. I took the idea brought about by Atlassian – the idea that once in a while, you have a day where a group of people can do whatever they want – hobbies, work, personal, family, whatever – with the single restriction that at the end of the day they shared what they did alongside an enjoyable, informal dinner. They discovered that this 20-80 ruled production and sharing period was mindblowingly useful. So useful until Google took up the same system and did similar with their employees.

So I took the attitude that whatever I did, I should be proud of it. There isn’t any use in doing something you aren’t proud of. I’m not proud of killing time, and so I made it known to people. I worked on what I loved. And then I shared what I loved. Sadly the feedback was less than favourable. People didn’t share the same interest I did in just hearing about things people love – irregardless of field or industry. Extrapolating that – I didn’t find people who wanted to be proud of what they did. They were content with just living. Perhaps I was searching in the wrong place, perhaps I was searching at the wrong time, perhaps I was searching for the wrong signals.

Slowly digesting this information – I realised more and more that WIPUP is built for almost nobody. It was designed for such a niche that the euphemism of the word “niche” (ie. most people simply don’t care) doesn’t apply any more. This brings up a very important thing to consider – what do I want to achieve for WIPUP outside my personal wants and needs?

In WIPUP’s current state, most of the previous users have moved on. Despite being online for a couple years, WIPUP is only home to 150 user accounts, only a handful of which are active (ie. can be counted on your hand), and of those which are active, the majority are people I have known for some time online. There are no advertisments, no total filesize restrictions, completely for free, and recently it seems as though some idiot has written a bot to register an account and insert updates with spam links in them. Development has stalled due to almost all of the features I wanted to include already implemented.

So what exactly, then, is WIPUP’s current objective?

I’m hunting, folks. I’m hunting. More to come.


WIPUP 27.06.10a released!

It’s super, it’s amazing, and it’s released. It’s WIPUP 27.06.10a. For the uninitiated, WIPUP is a flexible and easy way for people to share, critique, and track works-in-progresses.

To quote some random person, this release truly brings out the “hey, it’s like a working site now“. This release sports super fancy upgrades courtesy of my schedule, which is now free from exams and school. Check out the WIPUP website now, and read the release notes.

Of course it’s also open-source, so not only do we welcome new users, but developers too! This is hopefully the last “alpha” release, so feel free to join.

Life & much, much more

WIPUP now supports video updates!

Well the Eadrax code (what WIPUP runs on) has always supported updates with video attachments but the live WIPUP site never got to see it in action due to the server not having ffmpeg (video swiss-army knife) installed and playing nice with the latest codecs, permissions, and whatnot. Over the past few days our lovely host OpticEmpire has gotten ffmpeg up and running on the server – and it worked like a charm.

I uploaded a short clip I made a month or so ago to show the company Johnson Controls embracing the Generation Ys. It worked flawlessly – snapshotted a thumbnail halfway through the clip and resized it as necessary, reencoded the file into .flv format (HTML 5 and video tags are on the way folks, but meanwhile we have to keep legacy users happy), and the update page has a lovely in-browser video player courtesy of LongTail Video and their free license on JW Player.

The encoding is done on-the-fly but in the future encoding will be queued by the server so we don’t blow ’em up, and the JW Player will get skinned in a WIPUPish fashion. Note that videos will always be compressed. The point of WIPUP is to dump up unfinished works, and so at the moment it’s simply uneconomical to host uncompressed files.

Go and view the demonstrating update here.


WIPUP on Trac

Yesterday saw a burst of activity as far as WIPUP commits were concerned, but most importantly I’ve said goodbye to the proprietary Assembla project management tool we were using before and set-up our very own in-house service "Trac", courtesy of our lovely host.

Trac’s main use is in ticket tracking – a way for users to report bugs and wishlists they have with the website. However like any respectable project manager it also comes with a wiki, on which you can find the brand new Eadrax roadmap, so you can see what fancyness is coming up ahead. So if you’re playing with WIPUP already and see something you don’t think is quite right go check out trac.

Oh and if you missed the sneak peak showing BBCode and resizable forms, check it out now.


After the WIPUP release, the stats are in.

Unless you have short-term memory loss you would remember about a week ago WIPUP 14.01.10 was released. One of the major features of this release was a working and presentable stats section in the account dashboard. Generating statistics is a rather intensive operation on the webserver and would be overkill to have it happen daily, even when not considering how useless daily information would be in terms of updates (unless you give updates daily). So what happens is that every week it checks the stats against the database and redraws the images. It’s been a week since then, and even though WIPUP is a lonely corner of the internet at the moment, the numbers just don’t lie.

Yep, that’s the number of updates I’ve done per week. Last week was special, with two updates from me, but on average I only have 1 a week. Obviously I’m not dogfooding enough or I simply don’t have time for it. You’ll notice a correlation between when those updates appear with when we have a (WIPUP) prefixed update on thinkMoult.

This one draws a slightly different picture – but nevertheless still corresponds with when I release updates. When updates are released we see a surge in views. Interestingly during the two weeks when during both I released an update we see a slight drop, which may suggest that my first update was a bit disappointing (on 04/01). Of course with the major release we see a jump up around 200 views, which is quite nice I must say.

For the sake of completeness we’ve got to include this little guy – the comments chart. I’ve been rather dormant and not commenting on anything (well, nobody else uses it!) and likewise nobody has commented on my stuff (which is perfectly understandable, folks!). However after that major release thing I’m happy we attracted a kudos and a comment by a user named Clarkey. "Sexy", he says. (Personally I think that’s kudos material)

Let’s see how badly this drops down in the next week.


Dogfooding WIPUP

I’ve mentioned dogfooding a few times already: it is the process where the developer uses their own creations, thus duplicating the role as a consumer as well as a producer. It’s good practice – because if you don’t trust your own ideas, the chances are that it’s a bad idea in the first place. Even if you don’t use it simply because you are building it for a different personality profile, this implies that you are not truly in the customers’ shoes – how can you design something for somebody you know nothing about? More importantly, what motivates you behind your work, the quality of the work or the success it brings?

My ideals tell me that it should lie with the quality of the work. Ironic it should be that a while back I posted on the topic of WIPUP and webdesign usability – especially when WIPUP is a little online system that was literally built for my own use. A little closer inspection into that article shows that though it does give an example of bad webdesign usability, the point is that I had made an update and had something to show for it.

Yes, and that is the whole point of WIPUP.

If you didn’t get the hint by now, it means that that article should’ve never existed in the first place, but instead appeared on WIPUP. Luckily I realised this and submitted it – and here we have the total sum of my dogfooding, which isn’t much:

13 items – a shameful quantity at any rate if you look at how long I’ve had the opportunity to use WIPUP, unfinished as it is.

Anyway, it now exists and we’ve identified a few more shortcomings of the WIPUP update addition system:

  • There is no rich formatting for the detailed description of the update. I propose the addition of BBCode as well as a rich text formatting toolbar.
  • There is no way to extend the small textbox for writing a very long detailed description, making it inconvenient for the user.
  • There is no way to attach more than one file nor to make an anchored reference to an attachment.
  • There is no clear indication of which fields are required and which are optional except for a vague statement in the introductory paragraph.
  • Beforementioned introductory paragraph needs some rewording or rethinking, it’s utterly useless, and only adds to confuse the user.

See folks – dogfooding works! (in theory, let’s see how long it is until I actually update it). Meanwhile, why don’t you check out that update yourself.


My OpenDesktop Competition Submission: Wipup

Folks from PlanetKDE last heard me announcing my journey along the path to become a KDE developer. There are many ways to do this and unfortunately the path that involves learning a load of C++ and start developing applications is still making slow but steady progress and not (yet) eligible for public announcement.

But – there are many ways to contribute!

I knew about the OpenDesktop Competition for quite a while now and originating from the area of webdevelopment I realised that my latest project ties almost perfectly with its goals. Obviously being very much related to KDE development and open-source in general I wanted to share it here:

Click here to check out my submission.

Obviously the main way to make this project become successful is through community support. I really think this can be integrated well such as through plasmoids or plugins on applications such as Krita or Dolphin.

Sorry for not really explaining what it’s about because it’s quite difficult to explain very quickly. But here is a crappy attempt: It allows users and developers to showcase the works in progress of their projects and keep in touch through them.


Of course, if you like the idea, I would love feedback and voting :)


The Eadrax Dashboard

Project Eadrax, one of the main things I’ve been dedicating my time to nowadays is currently stalled. This is because to continue we really need a design. Unfortunately our designer promised more than he could deliver and it’s been a waiting game for quite a while now. He asked to be given one more week (until the 24th) and I really hope we have a design to present then. Worst case scenario we need another webdesigner, or at least somebody who’ll help me on icons and work with me on making the new design.

Meanwhile I wanted to showcase some of the new features we see in Eadrax that the old version didn’t have: the Eadrax Dashboard Charts. To be honest, this feature was actually implemented a few weeks back and were given the honours of having the second reported (and now fixed) bug. I was wondering what sort of statistics would make Eadrax an interesting system to use, at the same time subconciously persuading users to contribute content ;)


The first is the newsfeed. This is exactly like other systems, such as Facebook. It basically says what’s been going on with the people you are tracking and the projects you are subscribed to. It looks quite ugly now, as like I said before, we need a design. All that exists now the barebones system. If you’re wondering about the “asdf”-like text, that’s because I simply dumped in some data in the database to see if it worked – and yes, it did :) Of course, you might spot some errors here – for example how can I subscribe to my own project? How can I track myself? This is obviously impossible! Actually yes it is, but I just dumped in data to make it look realistic – it’s all fake.


The second are statistics about your own popularity. Who is tracking you and who are subscribed to your projects. It breaks it down per project for subscribes and we can probably display this information similar to how your “followers” are displayed on Twitter. Or we could be more original – ideas are welcome! So as you can see above, I have two projects, the “Eadrax” project and the “wefwef” project.


Now it gets more interesting. This shows how many updates you have submitted each week for the past 8 weeks. I simply added some dummy information and didn’t go too far back in time, so obviously when you actually start to use the website these graphs will look much, much more interesting. So according to this, I did nothing until the 6th of July, did a decent amount of work during that week, but slacked a bit during the next week.


Another graph shows how many comments I have given, and how many comments on my stuff I have received. We want to encourage feedback as an important part of this system, and we hope lots of comments (and constructive, too) turn up. Also, it’s quite fun to have two lines on one graph ;)


This graph is straight and to the point – it tells you if people are actually looking at your stuff. Obviously here the statistics don’t add up. For example, by the 13th of July I did not get any views, yet somebody commented on my stuff – how is that possible? This is because I made up values in the database just for testing purposes, of course – not because my code is buggy (crappy maybe, but I think it works)


Now we enter the lovely world of pie charts. This shows you how many people are subscribed to each project. Again, the math doesn’t add up – previously I had only mentioned two projects, the “Eadrax” and the “wefwef” project – so what is the “Trackers” project? In fact, I also mentioned that you can both “track” people and “subscribe” to individual projects. It therefore goes without saying that if you are tracking somebody, you are automatically subscribed to all of their projects. I hope that makes sense, and I hope that really makes this pie chart useful. For example, I can see that people really like the Eadrax project – perhaps I should update it more.


This pie chart shows kudos (the plural of kudos is kudos, whoops!) per project. If somebody likes what you did on an update, they can “kudos” it – this simply adds them up. As you can see here, some of our kudos belong in the “Uncategorised” section – that means not all updates need to be categorised within projects. This is useful if you are simply doing random testing or doodles that isn’t part of something major. What’s vital about this chart is that you can see it keeps the same colour codes, so you don’t get mixed up. Here we can see that even though it seems as though two people are notified when I update “wefwef” and only one (the person who is tracking me) is notified when I submit an uncategorised update, the uncategorised ones seem a lot more popular. Again, this is all just fake data :) I don’t really have a project called “wefwef”.


Finally we can see our own personal activity on our projects. As we can see here, even though I am equally active on both Eadrax and wefwef, Eadrax is significantly more popular as seen from our previous chart. I spend most of my time on random crap, but only one person out of a potential four people are notified about it – but rightly so that person seems to be a big fan of my random crap work, and so I get the most kudos on my uncategorised stuff. Useful, eh?

This data will probably be presented in some sort of grid layout once we have a proper user interface – but until then it’s just random images spewed here and there. However I wanted to show you what it’s looking like now and see if you like it!


Progress on Eadrax

Even though me blabbering about project Eadrax doesn’t seem to interest most people (see: 0 comments on previous 2 posts) and it seems as though nobody has responded (yet) on the mailing list, I do not consider it a total failure.

I’ve decided to share just how much data you can actually gather from taking a peek at the logs behind Eadrax.

Allow you me to indulge you in a recent graph from GitHub showing some logs of the progress on Eadrax:


For those unfamiliar with version control, think of it has a log of every single change that was made. Like a save file. Each dot over there represents an edit.

Those subscribed to the mailing list and that have read the Eadrax Space will have known that I have restarted Eadrax as an open-source project. The graph up there shows two people, ukd1 and Moult (me). ukd1 (Russel) is another developer I met in the #kohana IRC room, willing to join in the project. Before continuing, you cannot see this from the image, but just as a note, every single update done up there was done by me. The little tags represent “branches”, essentially different features/sections of the site that are broken down so that people don’t get confused what belongs where (and also so somebody doesn’t screw up everybody elses work). The master branch is the overall project (well, sort of).

From this you can garner some very interesting information:

1) Only Moult has done anything on the project so far.

This is not a bad thing–at early stages of the project getting so many people to create the underlying structure of the web application can lead to some very confusing situations. I also know that the other developers have been extremely busy with other things, but hopefully this will change in the near future.

2) Russel (ukd1) apparently decided to work on the project a while back, but then just stopped. His line is quite left behind with the recent changes.

Obviously he has some catching up to do.

3) When Russel last updated his project, only two branches existed, “master” and “users”, now there are many more.

This shows that the application structure is probably mature enough to start introducing new sections of  the site and thus more branches. This is a very good sign for progress.

4) The latest branch is no longer the “users” branch – in fact, it has been left behind.

Well, the users branch obviously has to do with the user system – logging in, registration, etc. This is very good proof that there is already a clear structure behind the system and now we can just build up upon it.

5) Recently the code has been splitting up and merging together.

Obviously I’ve been developing some things here, some things there, and it’s now good to start pulling things back together into a simple structure so other coders can know exactly where the code currently stands.

Summary: Well, nothing much interesting to be seen on the front end of things yet, but just a very good sign that things are moving along, and moving along quick. Expect by next week to have most of the code ported over from the previous Eadrax version.


Pre-alpha Eadrax Testing Begun!

…and we’ve begun (officially) the pre-alpha testing for Project Eadrax. All testers should have received several emails with relevant information. Actually, due to a slight muckup in the setup, you got spammed a little with one too many emails. Sorry, won’t happen again.

To give this post a bit more depth, I would also like to announce that I have upgraded to WordPress 2.8. Tada.

Might seem like nothing big has happened, but this is a significant benchmark that cannot be done justice by a blog post. What a lame excuse for today’s post ;)


Final call for Eadrax testers.

Hello–sorry for the rather short post today but exams are very nearly done with (Will be finished by Thursday afternoon!) and then Project Eadrax will start moving along full speed, and ThoughtScore will too be seeing some appearances. Yeah so as I’m quite busy no big post for today.

However, this is the final call for anybody who wants to join in on Eadrax testing. There are no prerequisites, just your own willingness, curiosity, and hopefully active involvement is wanted. You don’t need to know programming, design, or anything- though it would help if you did. This is for anybody and everybody.

In sharp contrast to the “anybody and everybody”, because of the early stage it’s at we don’t want to open it publicly. If you’re interested in joining, send me an email or leave a comment on this post with an email that you regularly check. You will then be notified with all the details later on :)

What is Eadrax? Well, it’s a website I’m working on. That’s enough for now!

Deadline is this Friday.