Wiki Blog

Dare to say what nobody dares to say

December 26th, 2009

Some times I come across a blog post that is so ad rem that I wish I had written it myself. One such blog post is "Thoughts on wikis" which you can find on the iperimeter blog.

From the article:

All wiki platforms I have used appear to make the following assumptions:
1) Formatting, beyond a very basic level, doesn't matter
2) People are prepared to put time and effort into learning and using the platform and/or its WYSIWYG editor
3) All content fits cleanly into a predefined structure
4) Once the information is in the wiki, people don't need to get it out again
5) Users are IT literate and have a conceptual interest in the wiki as a platform

I'd like add:

6) Users are able to figure out where to upload pictures so they can reference it using an Url

Of course, none of that is true. We are genuinely convinced that the true potential of wikis haven't even emerged. Once people can start to add content, move stuff around, do true WYSIWYG, all within the luxury of their familiar word processor, that's when we will start to see true content appearing.

Strangely, that is not the kind of comments you should make. Somehow, wikis still remain up to today a tool for the happy few. The technical thresholds seem to prevent many people from participating. Not a very democratic view. Along that same line, here is an other assumption that everybody seems to be making:

7) Wikis should all be open source and be free

That is true of course for the mother of all wikis: our beloved http://www.wikipedia.com. But even there, the money needs to come from somewhere.

We were even believing that for some time. Numerous people said that WordonWiki should really offer OpenOffice as an alternative editor. While that is certainly a feature on our backlog, interviews and usability sessions with real users have learned us that this is only a marginal request.

Let us know what you think.

Roland.

PS.
I was busy commenting this information on the author's blog, but some strange registration system was used. I had to login with my existing google or wordpress account. I did not feel comfortable doing that.

An unforgiving tale of extremes

October 25th, 2009

My world friends

They are many, Salman, Nagesh, Pavan, Matjaz, Umair, Masood and not to forget Vicky (who I secretly suspect is really a man). All have top level rails qualifications. Or so they say.

They are all part of the growing network of developers from the East. India, Pakistan, Slovenia, and so on.

If you’re considering outsourcing part of your product development, I can offer you some insights that were learned the hard way. To summarize, working with outsourcing partners is a story of extremes:

Things move amazingly fast in the outsourcing-sphere. If you post a project request on Friday night, you can easily get 10-15 project proposals by Monday morning. The majority will be so called bid spamming. More about that in a later post. But invariably there are 3 or more that are worth considering.

You need a ‘detailed’ specification. And I mean detailed. I usually deliver the mysql db creation script, a functional specification, a detailed design, a set of screen captures.

So what happens next blows you off your socks if you are not prepared for it. Among the big number of replies, there are a few replies that really strike you as ‘professional’. They have read the proposal, react to the hidden question you sneak in, often ask intriguing design or specification questions. So you reply and they surprise you yet again with their answers.

They do this for price less than the cost of the office cleaning lady.

Yes, Mister Frans
Of Course, Mister Frans

So the next step is to get to get to know who is replying. I would advice not to go for companies at first. There are many smart businessmen that have hired young developers eager for a job. They know what you want to hear:

• Yes Sir, we do complete quality control.
• Of course Sir, we send daily status report.
• Sir, this team is fully at your disposal. It is your team.
• We don’t stop Sir, until you are fully satisfied.
• If you have small change requests, don’t worry. We implement them for the same price.

Read this last line again. When was the last time you heard a Project Manager, American/European style, say that line. And you know what? It is true. My god, it is hard not to exploit that.

So go for the individual developer first. The guy that just left his company and now starts on his own. Often from his kitchen table. Strange coincidence, they invariably all worked at Microsoft. So test them. I mean, T.E.S.T. them.

Before I said you would need to check references (rentacoder.com keeps a nice history of user feedback). And not go below a score of 90% or even 95%. While that may help not to get bad ones, you would miss out on the real genius guys. For those ones, I cannot give you real rules. Trust your gut. Don’t feel pressured by time to select one early on in the process. I have kept some of my project requests open for weeks.

Now my limited sample clearly suggests that the bright individuals score the best scores. Two of the best experiences I had are with two individuals. On multiple occasions. As a consequence, I now give them jobs and get a result delivered in a few days, on one emergency situation, even in less than 12 hours.

The Girlfriend in China, or the motorcycle trip trough India.

Of course the individual has a life too. And he needs your money to pay for his plane ticket to China to meet his girlfriend. Or go motorcycling through India for two full months. And when they are not traveling, chances are high they are working on some other project, because other project owners have also found them.

Go for it. With caution

It is easy to lose your sense of professionalism. Because prices are so low, you tend to do a lousy job over specification. But trust me, my friend, when you are 3 time zones apart, clear, unambiguous specs matter even more. Some sites like rentacoder even encourage this. You can let two coders that bid for your project, do the job. Then pick the best one. I have yet to understand what that means and how (if at all) that would work.

My guess is: it does not work like that. You expect professional html, rails or dot net coding, the least you can do is be professional yourself. So spec to death, write acceptance scripts, and force yourself to have daily msn chats. Test, test and test. Give immediate feedback. Make them deliver the working system on your server and peek inside the code, the database and the logfiles. It is ok to be obsessive about it.

Good luck. You’ll need it.

Usability sessions: the online attempt

April 20th, 2009

We experienced already a number of times that usability testing is absolutely essential to create good software applications. If you haven’t bought Steven Krug’s “Don’t make me think‘ book, stop reading this post and order it. It’ll be your best buy this decade.

We are going through a redesign of our website and the screens. All we have is a static website, and a number of screen impressions (in PDF format). We have put up a question on a number of places. In this post, I’d like to share our experience about how that went.

“Linked in” is king

Of all places we have posted our invitation to validate our screens, LinkedIn.com was definitively the most successful. We got over 10 answers of people that really seem interested. “Yahoo Answers” was a distant second: only 1 answer on the 3^rd day. All other places (blog posts, develop forums…) gave 0 results. Zippo, nada, nieltsj (I just invented that one).

Free cookie

As part of our offer, we gave away free live time subscriptions. What did we notice?

• Nobody bothered to ask us what and how this FREE thing was going to happen. To be honest, we don’t yet know ourselves.
• When I asked those who participated if they really wanted the free subscription, they all confirmed.
• We were in no position to do a split test on this one, but from the reactions we got, I think ‘FREE’ at least drew some extra attention.

50% success rate

Of all people that initially voiced their wish to participate, only half actually replied. I didn’t push people to participate.

Be specific about the format

One lady said she was disappointed by the format of the test. She expected a functional prototype, not some static PDF thing. I guess we should have made this clearer at the start.

Magical answers

I monitored my inbox for days to get the feedback. When results started pouring in, we were truly amazed:

• the detail was phenomenal,
• the analysis was thorough and
• the similarities between all participants was striking.

Many people went over each screen in great detail, giving pro’s and con’s about various aspects of the design.

Conclusion

The online version of usability testing certainly misses out on some of the finer nuances of people’s reaction and initial confusion/enlightenment on your screens. However, when on the very early phases of design, I would compare this approach to the real live equivalent of ‘hallway testing‘. It gives you immediate feedback you just can’t ignore. So all participants certainly will get their FREE-Forever ticket.

Wiki switches to MSWord and Open Office

March 1st, 2009

This would probably make a ’scary’ headline for many. A recent post over at Simon Goh’s blog made an interesting comparison on how editing a wiki and authoring a document differs. Simon correctly argues that following MSWord features hinders a widespread wiki adoption:

• Too much options to indicate changes:
  • Comments
  • Revisions
  • Strikethrough
  • Highlighting
  • Any combination of the above
• No RSS
• A ‘document’ has an owner, so other people are less inclined to modify it.

I would add following topics:

• No easy ‘read’ feature
• No easy ’show history’ feature
• Cumbersome navigation between ‘wiki pages’, which could actually be ‘documents’.

Yet, wouldn’t we all want to combine the richer editing features with the community aspects of a Wiki. Of course, as we are building exactly this combination, we strongly believe in it.

Add comments on what stops MSWord users create the next Wikipedia.