Quality assurance
Filed under: technologies — andraz @ 17:08 24. Nov, 2007
How do you test a complex system that is trying to mimic being smart?
You want to automate testing, so you have your quality meter available for every little change you make. While having unit tests helps catching classical programming regressions, the major part of the challenge is having ’smart part’ under control. Unfortunately the only way to tell if the system is doing a good job or not is to have human check the results. The trick is that if you could automate testing in general, you would already have solved the hardest problem.So what you basically do is generate a set of evaluation data, manually. And have a system that does something like unit tests, but instead of giving you fail/pass results, you get statistics. Now you would think that the problem is solved, but that’s far away from truth.
There is changing of the dataset - when you have new content in the system, you get completely new related stories and you have to go back and have a human judge them. There is expansion of the evaluation data - as you add new tests you generally can’t send them through previous versions of your algorithms, since that would be prohibitely expansive. And there is statistics that hardly gives you overview over what exactly your changes caused, just few final numbers. And then there is the problem of pipelining the processing. Even if you improve the first stage, end results might be worse, since you’ve already adapted the second stage to previous first one. So you need to actually evaluate each part of the system in isolation and then together.
At the end you actually find out that you spend disproportional amount of time evaluating even the smallest changes. So you are in danger to just skip that evaluation which naturally you shouldn’t.
Ok, so much for today, now I think the evaluation run has just ended and I should be checking the results, again.
Related articles by Zemanta:
Linking in old media
Filed under: blogging experience, zemanta — andraz @ 2:42 23. Nov, 2007
How can you tell the difference between articles of big media and blogs?
Large majority of old-media that moved to internet has a very strict rules that they should never ever link to sites outside their conglomerate. If by some coincidence they do, they put all the links at the end of the article and make them as unnoticeable as possible.
The official line is that they are afraid to be held responsible for accuracy of the content they link to. The true reason is that either newspapers and TV-derived sites are trying to prevent their readers leaving or their journalistic staff simply hasn’t caught up with the meaning of the web yet.
I’ve seen a lot of both. From my time at RTV Slovenia I can tell that a lot of journalists (with exceptions!) haven’t grasped the idea of a reader/viewer being able to do basic research on their own. This is somehow excusable, big cultures are slow to absorb new concepts. There is not much that can be done (except waiting). However the first excuse - being afraid of losing readers - is really just an unexcusable strategic mistake.
Not providing to the audience direct links to referenced and relevant documents is actually a missed opportunity to fulfill the real need of curiosity. Missed opportunity to be user’s starting point of inquiry into any topic.
And then media companies complain how search engines are capturing all the users by leveraging other’s people content. Well, it is the media companies that are refusing to be the user’s home base which he can actually leave to explore the wilderness.
Related articles by Zemanta:
Evento Blog Espana
Filed under: blogging experience — andraz @ 2:34 22. Nov, 2007
This weekend Jure (a.k.a. the interface guy) is going to a Evento Blog Espana. He’s going to talk with spanish bloggers and gather some feedback from non-english speakers. If you [dear reader] also happen to be there, leave him a note, he might be in a mood for showing the alpha (jure at zemanta dot com).
Also he just updated me on the buzzwords we should be using when addressing tech community. So here it goes: we are BJAX!
All those nasty things we are doing with Firefox extension actually have their own name! BJAX means Browser Extensions and AJAX. The plugin does the following: when the page just loaded is identified as a “create new post” it injects a bunch of javascript that provides all the Zemanta functionality, plus some dirty details. This means people will be able to have Zemanta experience even when blogging on hosted platforms.
Isn’t that great?
How do you spell Zemanta?
Filed under: zemanta — andraz @ 1:35 21. Nov, 2007
Some internal jokes that really aren’t:
Q: How do you know you are stretching your tools to the limit?
A: You find reproducable segfault inside Perl. In its regex code.
Q: How do you know you surely have enough RAM?
A: “free” command causes a line break in a terminal. Everything below is doubtful.
Q: How do you know your server specs are not balanced?
A: You have more memory than disk space.
Someone testing one of the most hyped technologies in the Valley: Who are “The other guys” and where can I use them?
Q: What does a developer say to CEO at 8am in the morning?
A: Goodnight.
Q: How many developers does it take to change a lightbulb?
A: Wait, wait, I am already looking it up in Wikipedia!
Q: How do you properly spell Zemanta?
A: Did you mean Zemanova? Google cannot be wrong!
Related articles by Zemanta:
Mobile broadband in UK
Filed under: london, seedcamp — andraz @ 12:10 29. Sep, 2007
Getting mobile internet in London is really hard. Last few days we went to all major operators and asked them a simple question: We need mobile broadband. Can you help us?
After telling them that we don’t have UK bank account yet they were baffled. We explained that we need mobile broadband for three months and are willing to pay (almost) any price under condition that we get it right away. Even telling them that we are willing to pay one year subscription fee in advance didn’t help. They did not want our money.
The funny thing is Vodafone actually has a plan where you have no monthly fee and pay just for the traffic. We wanted to pay for traffic in advance, no go. They still wanted 18 month contract, even though there is no monthly fee.
The story has a happy ending, though. A fellow Seedcamper helped us taking a contract on his name. We are sincerely grateful. And he’ll also help us to make sense with our personal money.That is when we actually get UK bank account.
Update: Zemified!