"Code Sandwiches"

Daniel Gibson metalcaedes at gmail.com
Wed Mar 9 14:56:08 PST 2011


Am 09.03.2011 23:38, schrieb Nick Sabalausky:
> "Daniel Gibson" <metalcaedes at gmail.com> wrote in message 
> news:il8t79$2t70$2 at digitalmars.com...
>> Am 09.03.2011 22:49, schrieb Daniel Gibson:
>>> Am 09.03.2011 22:33, schrieb Nick Sabalausky:
>>>> "Nick Sabalausky" <a at a.a> wrote in message
>>>> news:il8rmg$176i$1 at digitalmars.com...
>>>>>
>>>>> But why is it that academic authors have a chronic inability to release
>>>>> any form of text without first cramming it into a goddamn PDF of all
>>>>> things?
>>>>
>>>> It's like how my dad tries to email photos by sticking them into a Word
>>>> document first. WTF's the point?
>>>>
>>>
>>> No it's not.
>>> At least PDF is a standard format with free and open viewers on about any 
>>> platform.
> 
> Vaguely free, open and standard. Only in the same sense that swf, doc and 
> docx are free, open and standard. HTML (bad as it may be) still wins here.
> 

No, PDF is an ISO standard, swf and doc aren't and docx isn't either, because it
doesn't really conform with the OOXML ISO standard..

As mentioned before: there are free and open viewers for PDF for (almost?) all
platforms that work reasonably well.
Can't say the same about doc(x) or swf..
That HTML is rendered almost the same on different browsers is a pretty recent
development as well...
Nevertheless HTML doesn't have as much formatting possibilities as LaTeX,
especially for formulas, so you'd end up using a lot of images which is suboptimal.
(Yeah I know there's MathML, but AFAIK it's not properly supported by all browsers).

> 
>>> And while sticking photos into a Word document is pretty pointless using 
>>> PDF for
>>> papers does make sense.
>>>
>>> One thing is that papers are usually published in printed form,
> 
> Still?
> 

I think so. And even if they aren't they're formatted like that anyway.

> 
>>> the PDFs are
>>> more or less a by-product of that.
>>> Also they're usually written with LaTeX (or something similar) and the 
>>> obvious
>>> (digital) formats to publish stuff written in *TeX are Postscript and 
>>> PDF - I
>>> guess you agree that PDF is preferable, as it can be searched etc ;)
> 
> *Some* PDFs can be searched.
> 

Most can, the others are - most probably deliberately - broken.
You can do the same with HTML if you want, just use images instead of real text..

> 
>>> You can also export *TeX to HTML, but that'll probably fuck up formatting 
>>> and
>>> formulas. So you'd have to use some LaTeX->HTML converter and clean up 
>>> stuff
>>> afterwards to make sure the formatting is OK, the formulas are like they 
>>> were
>>> intended to be (missing a small detail like a ' or an index or whatever 
>>> will
>>> make a formula unusable) etc..
> 
> So after 15 years there still isn't a good Latex->HTML converter? Sounds 
> more like the matter is a lack of interest in using anything other than PDF 
> rather than a lack of a good Latex->HTML converter.
> 

I don't know. I think I don't have to tell someone who still uses Firefox2 that
people don't have the motivation to try new software all the time just because
it may finally be usable ;)

> 
>>> This may not be a problem for this specific paper (it's only text, 
>>> sourcecode
>>> and some tables I think), but for many other scientific papers it is.
>>> That's the reason why they're mostly published as PDFs.
>>>
>>> Cheers,
>>>
>>> - Daniel
>>
>> One more thing: Published papers will probably be cited by other papers or
>> theses. With PDF this is easier, you can write "XYZ, page 42, l 13" - with 
>> HTML
>> pages it's not that easy, you could maybe write "in chapter 3 somewhere in 
>> the
>> 5th paragraph" or something like that, but that sucks.
>> Or worse "on the fourth page in the third paragraph" and once a new CMS is 
>> used
>> that splits pages differently that is completely meaningless..
> 
> These formal papers are divided into sections and subsections, plus HTML 
> supports links and anchors, and even supports disabled word wrapping if 
> that's really needed, so those are non-issues.
> 

If anchors etc are used.. fine. But you can't take that for granted.


More information about the Digitalmars-d mailing list