This site is almost completely unreadable without cleartype (windows XP/Opera10)
I hate cleartype, it makes everything look slightly blurred - it’s bad enough that everything in software is slightly blurred without the text looking like that.
mgb, uninstall the C fonts (typically installed with Vista or Office 2007) if you don’t want ClearType. They’re designed for ClearType and will never look good on any system without ClearType enabled.
The stylesheet defines fallback fonts but you aren’t seeing them because you have these C fonts installed.
@mgb: just fix your ClearType settings using the Microsoft Clear Type Tuner:
http://www.microsoft.com/typography/cleartype/tuner/step1.aspx
It doesn’t look blurry if you have it set up right.
The history flow visualisation is quite interesting.
I assume it breaks down when large sections of text are being moved in an article though?
(e.g. if I decide to re-order the sections of an article, without re-wording any of it, then I am actually performing a fairly minor edit - but it would look massive on the history flow)
Nice idea, but I question the reasoning behind it.
If you knew this question was from Turing Award winning computer scientist Alan Kay, would it change the way you reacted to it? Of course it would!
No, it wouldn’t. A good question is a good question, and whether or not I answer it is not going to be influenced by who wrote it. The only exception I can think of is if I personally know the author - but squealing fanboyism for someone famous isn’t going to play a part.
Put it another way - if Alan Kay asks a question that I have no interest in answering, I’m not going to change my mind just because it’s Alan Kay. If Joe Bloggs from Mundaneshire asks a really interesting question, I’m not going to ignore it just because it’s Joe Bloggs from Mundaneshire.
I knew before you told us that the post was from Alan Kay…because he signed his name at the bottom.
Thanks Jeff - I had to boot into safe mode to remove them/don’t you just love .msi.
Graham - thanks that did help a little. But Proggyclean looks bad in visual studio, I’m trying a few cleartype programming fonts.
Put it another way - if Alan Kay asks a question that I have no interest in answering, I’m not going to change my mind just because it’s Alan Kay. If Joe Bloggs from Mundaneshire asks a really interesting question, I’m not going to ignore it just because it’s Joe Bloggs from Mundaneshire.
That may be true, but important and powerful ideas probably means something different to Alan Kay than it does to Joe Bloggs. The identity of the asker has the potential to, in effect, change the question being asked.
While I agree with the thought behind the article, that knowing who the author is is a good thing… I have to ask… does it matter when the chances that you actually recognize the autior are slim to none?
If Wikipedia told you that CoolKid21 contributed majority of the content instead of last editor, Hottie84, would it really matter? Or even if they have real names… Doesn’t matter, I would never recognize any of them, so it would all just be the same to me.
Did Alan Kay ever get a satisfactory answer to his question? I like how he refuted most of the responses usually stating that a given invention was already invented at Xerox PARC in the 70s.
Why is it important to know who wrote/edited what?
If the answer/question/comment is good, I’m going to vote it up no matter who wrote it or who is the current owner of the message. If you react differently depending on who wrote something, maybe you’d be better off not knowing so you can decide by yourself if the information is good or not.
The only purpose of having access to the author is to go read more about him/her in its profile in case he/she said something meaningful.
@mgb: Consolas is easily the best ClearType programming font I have ever used.
It was developed by Microsoft specifically for programming and it is very clear and easy to read.
Jeff did an article a while ago about it:
http://www.codinghorror.com/blog/archives/000969.html
http://www.codinghorror.com/blog/archives/000356.html
Why is it important to know who wrote/edited what?
It might tell you if the answer is likely to be correct.
It is difficult on a forum to establish level of knowledge, the SO rep does this to some extent other forums have badges for long standing members.
Great work done here Jeff to work out a wiki ownership but doesnt that completely undermines the purpose.
No doubt it will influence readership and people’s reactions BUT when we say something is a ‘community wiki’, it means its been written by the community and the highest contributor (even if he is the only contributor) does it altruistically.
So it comes down to altruism or egoism… being a common face in the community or illuminated by limelight.
–A food for thought–
Is it still relevant that i contributed 90% of the lines if someone just changed the entire point i stated by chaning one of the 10 lines, and it is still endorsed by my name?
A very slightly related problem: as part of a system for automatically assigning crash bugs to engineers for investigation, I want to establish an ‘owner’ for each source file in a large code base. My solution: for each change to the file in the version control database, score N points where N is the revision number of the change. Thus more recent changes are weighted higher, but if person A creates the file (revision 1 for 1 point), person B makes two changes (rev 2 and 3 for 5 points), and person C makes the latest change (rev 4 for 4 points), B is the file owner. The script that computes the owner can also tell you the top-N owners; in this case it would say B 50%, C 40%, A 10%.
We applied this to a codebase inherited from an outside source, so weighting the initial checkins low makes sense (the day-1 import of 10000 files to the source control system wasn’t a creative act), but newly created files might ought to get a bonus for first checkin.
How do you get those graphs from wikipedia? (I know it’s in discover, but if this is a wikipedia tool it would be fascinating)
Nice new feature, too bad it’s broken.
If you read the original post by Alan Kay and the current revision you’ll notice that the text is identical. The only revisions made were a couple of re-tagging and making one work into a link. I’d still say that 100% of the text there is written by Alan Kay.
@Hamilton-Lovecraft Neat idea. If your language supports exceptions, then you can probably find out who last modified the line/function/file of the function calls to do some more accurate scoring based on the code that actually generated the error.
This post makes me wonder what diffing algorithms are used by the various source control systems out there.
I think the fact that most of the contributions to wikipedia are unregistered users may reflect badly on stackoverflow. Since you need to pass a pretty high bar in order to edit posts (on the order of registering and being active for several weeks) you miss out on most of the brain power.