View mode: basic / threaded / horizontal-split · Log in · Help
January 22, 2013
Re: D popularity
On 1/22/13 12:19 AM, H. S. Teoh wrote:
> On Mon, Jan 21, 2013 at 10:57:39PM -0500, Andrei Alexandrescu wrote:
>> On 1/21/13 10:32 PM, Nick Sabalausky wrote:
>>> On Mon, 21 Jan 2013 21:03:55 -0500
>>> Andrei Alexandrescu<SeeWebsiteForEmail@erdani.org>   wrote:
>>>
>>>> On 1/21/13 3:06 PM, H. S. Teoh wrote:
>>>>> I have to admit that ddoc hasn't really done it for me (yet). I
>>>>> think the impediment is that the default output needs some work
>>>>> (mainly in CSS and indexing) before it is useful as actual
>>>>> documentation.
>>>>
>>>> We should make a few macro batteries (HTML, mobi, pdf) readily
>>>> available in e.g. tools/ or examples/.
>>>>
>>>
>>> Is that possible? PDF is binary, but I thought DDoc only did text
>>> output? Come to think of it, how *do* the non-html versions of the
>>> docs currently get generated?
>>
>> LaTeX.
> [...]
>
> Speaking of which, did you manage to polish up the LaTeX macros to
> produce decent output? I'd love to take a look if you did -- I'm a LaTeX
> fan.

Yah, a start is already on github, run

make VERSION=2.061 `pwd`/web/dlangspec.pdf

from the d-programming-language.org repo.


Andrei
January 22, 2013
Re: D popularity
On Tuesday, 22 January 2013 at 00:32:03 UTC, John Colvin wrote:
> On Tuesday, 22 January 2013 at 00:24:04 UTC, John Colvin wrote:
>
>> P.S.
>> dmd 32bit on linux produces functional shared libraries, which 
>> can be happily called by IDL
>
> to clarify: extern(C) is required, there's nothing clever going 
> on.

Another correction: dmd can't make the runtime work. Ldc does 
work however.
January 22, 2013
Re: D popularity
On Monday, 21 January 2013 at 21:27:54 UTC, Nick Sabalausky wrote:
> If text editors written in JavaScript have become
> commonplace (<sarcasm>Thanks, Google!</sarcasm>), I'm sure 
> JS-based
> interpreters, JS-based codecs and "F"FTs (rather SFTs), and 
> other
> such nonsense aren't far behind.
You probably already saw these, but an x86 emulator that runs 
linux, written in javascript:

http://www.geek.com/articles/chips/javascript-emulator-lets-linux-run-in-a-browser-tab-20110517/

Forge, a Transport Layer Security implementation written in 
javascript:

http://digitalbazaar.com/2010/07/20/javascript-tls-1/

Not quite FFTs or codecs, but not far off. :) I think these are 
horribly dumb ideas, just pointing out that they exist.

> A roundabout way to say it, but I guess the point I started out 
> trying to make is this: The popularity of 
> dynamic/interpreted/sandboxed/etc
> languages *is* IMO one of the more significant roadblocks in 
> the way of D popularity. Silent fire alarms are what's hip, and 
> here we are
> peddling an old-fashioned sounds-and-lights fire alarm. We're 
> pragmatic instead of cool.

I agree with this generally, but I'll note that those dynamic 
languages really aren't that popular.  Using TIOBE data, (yes, 
yes, I know it's not good data, but I'm going to keep looking 
under the streetlight, dammit ;) )

http://www.tiobe.com/index.php/content/paperinfo/tpci/index.html

Three out of the top four languages are now native compiled, with 
C even beating back Java for the top spot recently.  D is 
probably in the top 7 for compiled languages and the only one to 
be designed in the last decade or so (I don't think Pascal and 
Ada are coming back ;) ).  So I think D is well-positioned to hit 
that top spot, or at least that's what I told a friend recently 
when I explained why I'm trying to use D. :)
January 22, 2013
Re: D popularity
On 1/21/13 11:22 PM, Knud Soerensen wrote:
> == online documentation ==
> When I returned to D I felt very frustrated with D online documentation.
> After looking for root of the feeling i found that the reason was that
> I was used to the documentation of php.
> In php each function have a page of it own.

We need to get on this, pronto. Unfortunately Sönke Ludwig, who had a 
working prototype, doesn't have the time to work on it at this time. 
Anyone would want to pick up the torch?

Just created http://d.puremagic.com/issues/show_bug.cgi?id=9370


Andrei
January 22, 2013
Re: D popularity
On 1/22/13 4:46 AM, Jacob Carlborg wrote:
> On 2013-01-22 04:57, Andrei Alexandrescu wrote:
>
>> LaTeX.
>
> Isn't that a failure of ddoc. Wasn't ddoc supposed to be independent of
> the output format. If we're only using ddoc to generate the HTML
> documentation it seems it's not good enough.
>
> This is not a big issue. I mostly care about the HTML docs.

Not sure I understand. We have working templates that produce passable 
PDF via LaTeX.

Andrei
January 23, 2013
Re: D popularity
On 2013-01-22 20:47, Andrei Alexandrescu wrote:

> Not sure I understand. We have working templates that produce passable
> PDF via LaTeX.

It's somewhat cheating to use an intermediate language like LaTeX, but 
in the end I guess it really doesn't matter.

-- 
/Jacob Carlborg
January 23, 2013
Re: D popularity
On Wednesday, 23 January 2013 at 07:18:37 UTC, Jacob Carlborg 
wrote:
> On 2013-01-22 20:47, Andrei Alexandrescu wrote:
>
>> Not sure I understand. We have working templates that produce 
>> passable
>> PDF via LaTeX.
>
> It's somewhat cheating to use an intermediate language like 
> LaTeX, but in the end I guess it really doesn't matter.


yeah...
in the end, it doesn't even matter.
http://en.wikipedia.org/wiki/In_the_End
January 23, 2013
Re: D popularity
On 1/23/13 2:18 AM, Jacob Carlborg wrote:
> On 2013-01-22 20:47, Andrei Alexandrescu wrote:
>
>> Not sure I understand. We have working templates that produce passable
>> PDF via LaTeX.
>
> It's somewhat cheating to use an intermediate language like LaTeX, but
> in the end I guess it really doesn't matter.

No, it's not cheating at all. You do realize the enormous effort that 
would be involved in producing good quality typesetting?

Andrei
January 23, 2013
Re: D popularity
On 1/22/2013 11:53 PM, Andrei Alexandrescu wrote:
> On 1/23/13 2:18 AM, Jacob Carlborg wrote:
>> On 2013-01-22 20:47, Andrei Alexandrescu wrote:
>>
>>> Not sure I understand. We have working templates that produce passable
>>> PDF via LaTeX.
>>
>> It's somewhat cheating to use an intermediate language like LaTeX, but
>> in the end I guess it really doesn't matter.
>
> No, it's not cheating at all. You do realize the enormous effort that would be
> involved in producing good quality typesetting?

Also, Ddoc's charter is not to generate PDF files. It is to provide enough macro 
capability that Ddoc's output could be, given the right macro definitions, fed 
into a PDF generating tool or any other reasonable document format.

The default set of macros will generate basic html.
January 23, 2013
Re: D popularity
On 1/21/13 10:06 PM, H. S. Teoh wrote:
> On Mon, Jan 21, 2013 at 09:03:55PM -0500, Andrei Alexandrescu wrote:
>> On 1/21/13 3:06 PM, H. S. Teoh wrote:
>>> I have to admit that ddoc hasn't really done it for me (yet). I think
>>> the impediment is that the default output needs some work (mainly in
>>> CSS and indexing) before it is useful as actual documentation.
>>
>> We should make a few macro batteries (HTML, mobi, pdf) readily
>> available in e.g. tools/ or examples/.
> [...]
>
> +1. That would lower the "barrier to entry" and make it more likely for
> people to adopt ddoc.

http://d.puremagic.com/issues/show_bug.cgi?id=9379

Preapproved, too! This may be an awesome website project.


Andrei
5 6 7 8 9 10
Top | Discussion index | About this forum | D home