Extension syntax vote begins - Wikipedia-l - lists.wikimedia.org

List overview All Threads
Download

Extension syntax vote begins

question to Andre Engels

Re: [Wikipedia-l] Re: Languages:...

Erik Zachte

6 Apr 2004 6 Apr '04

2:30 a.m.

Several people complained about this vote being too technical. Actually the vote is mainly about usability. What syntax would you find easiest to use or even only to recognize when editing an article. Nothing technical about that. Erik Zachte

Reply

Show replies by date

Timwi

6 Apr 6 Apr

9:51 a.m.

Erik Zachte wrote:

Several people complained about this vote being too technical. Actually the vote is mainly about usability. What syntax would you find easiest to use or even only to recognize when editing an article. Nothing technical about that.

That is true. Nevertheless, it seems that so far mostly technically-minded people have voted. I suppose the less technically-minded people (a) don't really quite get what this is about (b) don't care enough to read the huge bulk of arguments for and against everything. It is not uncommon for less technically-minded people to stay out of this kind of thing and blindly trust the "higher-ups". It is also not uncommon for them to have no real idea which option they will like better once it goes into actual production use. In fact, most people probably just take things like <math> for granted the way they are and aren't really consciously aware of the fact that they can influence it. The vote therefore has a systemic bias towards the programmers, but they are not the intended audience. Also, most of the programmers (apparently) vote on what they prefer (as opposed to what they think unsavvy people might prefer). There is no other way I can imagine that people seriously prefer <math>x^2</math> over simple-and-quick [!x^2!] or [$x^2$] or whatever. Except for <rend type="math">, *all* proposed syntaxes are better than <math>. Timwi

Reply

Arvind Narayanan

5:22 p.m.

On Tue, Apr 06, 2004 at 10:51:12AM +0100, Timwi wrote:

The vote therefore has a systemic bias towards the programmers, but they are not the intended audience. Also, most of the programmers (apparently) vote on what they prefer (as opposed to what they think unsavvy people might prefer). There is no other way I can imagine that people seriously prefer <math>x^2</math> over simple-and-quick [!x^2!] or [$x^2$] or whatever.

I think its just the opposite. Only perl programmers would prefer so many special characters rather than words to delimit things. Most people are familiar with html these days, so "unsavvy" people won't be frightened by <math></math> Arvind

Except for <rend type="math">, *all* proposed syntaxes are better than <math>.

-- Its all GNU to me

Reply

Tomasz Wegrzanowski

11:37 p.m.

On Tue, Apr 06, 2004 at 10:52:32PM +0530, Arvind Narayanan wrote:

On Tue, Apr 06, 2004 at 10:51:12AM +0100, Timwi wrote:

The vote therefore has a systemic bias towards the programmers, but they are not the intended audience. Also, most of the programmers (apparently) vote on what they prefer (as opposed to what they think unsavvy people might prefer). There is no other way I can imagine that people seriously prefer <math>x^2</math> over simple-and-quick [!x^2!] or [$x^2$] or whatever.

I think its just the opposite. Only perl programmers would prefer so many special characters rather than words to delimit things.

I'm a Perl programmer and I prefer <math>.

Reply

Peter Gervai

7 Apr 7 Apr

8:07 a.m.

On Tue, Apr 06, 2004 at 10:52:32PM +0530, Arvind Narayanan wrote:

On Tue, Apr 06, 2004 at 10:51:12AM +0100, Timwi wrote:

The vote therefore has a systemic bias towards the programmers, but they are not the intended audience. Also, most of the programmers (apparently) vote on what they prefer (as opposed to what they think unsavvy people might prefer). There is no other way I can imagine that people seriously prefer <math>x^2</math> over simple-and-quick [!x^2!] or [$x^2$] or whatever.

I think its just the opposite. Only perl programmers would prefer so many special characters rather than words to delimit things. Most people are familiar with html these days, so "unsavvy" people won't be frightened by <math></math>

I am a perl programmer and I disprefer [# .. #]. I could probably memorize them, but I don't really want to, and I don't believe laypersons could. If it reads "math" then it is obvious that what it is, you don't have to memorize. Test: <movie>...</movie> Try to guess what kind of media is handled by this imaginary tag. ;-) That's why. grin

Reply

David Friedland

6 Apr 6 Apr

6:06 p.m.

Timwi wrote:

There is no other way I can imagine that people seriously prefer <math>x^2</math> over simple-and-quick [!x^2!] or [$x^2$] or whatever. Except for <rend type="math">, *all* proposed syntaxes are better than <math>.

Did you really want to bring this debate onto the list (again)?! The simple answer is that <math> </math> is self-documenting. I look at [! !] and say "what does that mean?" and I look at <math> </math> and say "oh that must have something to do with math." The reason that non-savvy users might prefer <math>x^2</math> is that it requires understanding the concept of markup, a concept which most people learn in the context of HTML, and the very first thing you learn when you attempt to learn HTML is that things are enclosed in things that look like <something> </something>. That's why <math></math> is preferable to something like [! !], because it looks to most people what their idea of "markup" is. So even if a user knows only one thing about how HTML works, chances are it'll be enough for them to intuitively understand <math></math>. On the other hand, although [! !] is simple and/or quick, it is intuitive to no one. - David

Reply

Timwi

7 Apr 7 Apr

11:57 p.m.

David Friedland wrote:

Timwi wrote:

There is no other way I can imagine that people seriously prefer <math>x^2</math> over simple-and-quick [!x^2!] or [$x^2$] or whatever. Except for <rend type="math">, *all* proposed syntaxes are better than <math>.

The reason that non-savvy users might prefer <math>x^2</math> is that it requires understanding the concept of markup, a concept which most people learn in the context of HTML, and the very first thing you learn when you attempt to learn HTML is that things are enclosed in things that look like <something> </something>.

Which is also why we use <b>...</b> instead of ''', <a> instead of [[...]], <li> instead of *, etc.? Timwi

Reply

Peter Gervai

8 Apr 8 Apr

6:26 a.m.

On Thu, Apr 08, 2004 at 12:57:50AM +0100, Timwi wrote:

David Friedland wrote:

Timwi wrote:

There is no other way I can imagine that people seriously prefer <math>x^2</math> over simple-and-quick [!x^2!] or [$x^2$] or whatever. Except for <rend type="math">, *all* proposed syntaxes are better than <math>.

The reason that non-savvy users might prefer <math>x^2</math> is that it requires understanding the concept of markup, a concept which most people learn in the context of HTML, and the very first thing you learn when you attempt to learn HTML is that things are enclosed in things that look like <something> </something>.

Which is also why we use <b>...</b> instead of ''', <a> instead of [[...]], <li> instead of *, etc.?

If we would use math as often as ''this'' then I would definitely vote against long tags. But we use them rarely, and it's good that they're more descriptive that way. Often used contructs shall be short because we use them every day. (Offtopic: as far as I remember '' isn't equal to <i> but <em>.) grin

Reply

Ivo Köthnig

11:04 a.m.

Am Donnerstag, 8. April 2004 08:26 schrieb Peter Gervai:

On Thu, Apr 08, 2004 at 12:57:50AM +0100, Timwi wrote:

David Friedland wrote:

Timwi wrote:

There is no other way I can imagine that people seriously prefer <math>x^2</math> over simple-and-quick [!x^2!] or [$x^2$] or whatever. Except for <rend type="math">, *all* proposed syntaxes are better than <math>.

The reason that non-savvy users might prefer <math>x^2</math> is that it requires understanding the concept of markup, a concept which most people learn in the context of HTML, and the very first thing you learn when you attempt to learn HTML is that things are enclosed in things that look like <something> </something>.

Which is also why we use <b>...</b> instead of ''', <a> instead of [[...]], <li> instead of *, etc.?

If we would use math as often as ''this'' then I would definitely vote against long tags. But we use them rarely, and it's good that they're more descriptive that way. Often used contructs shall be short because we use them every day.

If we use them rarely or not depends on the article. Good math-article needs often references to math-objects by using variable-names. The clean way would be to set these variable always in math-markup and in this case we would need a short syntax for such articles. But I guess this is the main-problem. We are writing an encyclopdia about anything and so we need markup vor anything. But markup vor anything can not be short. --Ivo Köthnig

Reply

Peter Jaros

9 Apr 9 Apr

2:07 a.m.

On Apr 7, 2004, at 7:57 PM, Timwi wrote:

David Friedland wrote:

Timwi wrote:

There is no other way I can imagine that people seriously prefer <math>x^2</math> over simple-and-quick [!x^2!] or [$x^2$] or whatever. Except for <rend type="math">, *all* proposed syntaxes are better than <math>.

The reason that non-savvy users might prefer <math>x^2</math> is that it requires understanding the concept of markup, a concept which most people learn in the context of HTML, and the very first thing you learn when you attempt to learn HTML is that things are enclosed in things that look like <something> </something>.

Which is also why we use <b>...</b> instead of ''', <a> instead of [[...]], <li> instead of *, etc.?

Think about it: '' and ''' make a lot of sense visually. Same goes for [[...]], at least more sense than <a href=...>. Why <a>!? (*I* know why, but many people don't.) And it's *really* obvious that * and # are easy to interpret. These symbols make the syntax *more* human-readable. For extensions it starts to get muckier and, after all, less common. Myself, I'd like to have the XML-like tags for inline content and something more {{msg:xyz}}-like for transclusion. That, to me, is the most readable solution. Peter -- ---<>--- -- A house without walls cannot fall. Help build the world's largest encyclopedia at Wikipedia.org -- ---<>--- --

Reply

7343

days inactive

7346

days old

wikipedia-l@lists.wikimedia.org

Manage subscription

9 comments

8 participants

tags (0)

participants (8)

Arvind Narayanan
David Friedland
Erik Zachte
Ivo Köthnig
Peter Gervai
Peter Jaros
Timwi
Tomasz Wegrzanowski