Return Styles: Pseud0ch, Terminal, Valhalla, NES, Geocities, Blue Moon.

Pages: 1-4041-

If you think XML is bad

Name: Anonymous 2007-10-30 8:36

Then consider this, as it could be a lot worse:

UNA:+.? 'UNB+UNOC:2+STUB+BLA+960209:0843+72'UNH+0000090001+ORDERS:D:93A:UN:EAN007'BGM+220+B00404'DTM+137:19960209:102'NAD+BY+++STADT- UND UNIVERSITAETSBIBLIOTHEK:FRANKFURT+BOCKENHEIMER LANDSTR. 134-138+FRANKFURT+++DE'RFF+API:DE114110388'RFF+IT:STUB'NAD+SU+++B.H.BLACKWELL'CUX+2:GBP:9'LIN+1'PIA+5+0471949086:IB'QTY+21:1'FTX+ACB+3++TEST EDILIBE'PRQ+YYY:37.5:CA:SRP'RFF+QNB:00023302:10'DTM+171:19960208:102'RFF+LI:8216'RFF+BFN:S.KON.41'LIN+2'PIA+5+0471949000:IB'IMD+F+010+:::Cherry'IMD+F+011+:::Gordon E'IMD+F+050+:::Birmingham'IMD+F+060+:::a study in geography, hislanning'IMD+F+110+:::Chichester'IMD+F+120+:::Wiley(John)(W Sussex)'IMD+F+170+:::1994'IMD+F+180+:::254p'IMD+F+181+:::?: ill ; 24cm. - Bibl.?: P.237-244.'IMD+F+270+:::39100200?:Urban studies'QTY+21:1'FTX+ACB+3++TEST EDILIBE'PRI+YYY:37.5:CA:SRP'RFF+QNB:00023302:9'DTM+171:19960208:102'RFF+LI:8217'RFF+BFN:S.KON.39'LIN+3'PIA+5+0471957542:IB'QTY+21:1'FTX+ACB+3++TEST EDILIBE'PRI+YYY:45:CA:SRP'RFF+QNB:00023302:11'DTM+171:19960208:102'RFF+LI:8218'RFF+BFN:S.KON.41'LIN+4'PIA+5+0521435145:IB'QTY+21:1'FTX+ACB+3++TEST EDILIBE'PRI+YYY:65:CA:SRP'RFF+QNB:00023302:13'DTM+171:19960208:102'RFF+LI:8220'RFF+BFN:S.KON.07'LIN+5'PIA+5+0077090632:IB'IMD+F+010+:::Wenlock'IMD+F+011+:::Alison'IMD+F+050+:::The AS/400'IMD+F+060+:::built for business'IMD+F+110+:::London'IMD+F+120+:::McGraw-Hill Book Co in Europe'IMD+F+170+:::1995'IMD+F+180+:::134p'IMD+F+181+:::; 25cm. - Index'IMD+F+190+:::The IBM McGraw-Hill series'IMD+F+270+:::61100200?:Computers'IMD+F+300+:::1st Series Id?:E1678043'QTY+21:1'FTX+ACB+3++TEST EDILIBE'PRI+YYY:24.95:CA:SRP'RFF+QNB:00023302:1'DTM+171:19960208:102'RFF+LI:8221'RFF+BFN:S.KON.41'LIN+6'PIA+5+0521474620:IB'QTY+21:1'FTX+ACB+3++TEST EDILIBE'PRI+YYY:65:CA:SRP'RFF+QNB:00023302:14'DTM+171:19960208:102'RFF+LI:8222'RFF+BFN:S.KON.41'LIN+7'PIA+5+0133361993:IB'IMD+F+010+:::Shepperd'IMD+F+011+:::Martin'IMD+F+050+:::Foundations of software measurement'IMD+F+110+:::Hemel Hempstead'IMD+F+120+:::Prentice-Hall Interl Inc'IMD+F+170+:::1995'IMD+F+180+:::234p'IMD+F+181+:::; 24cm. - Bibls. - Index'IMD+F+270+:::61305050?:Software engineering'QTY+21:1'FTX+ACB+3++TEST EDILIBE'PRI+YYY:19.95:CA:SRP'RFF+QNB:00023302:3'DTM+171:19960208:102'RFF+LI:8224'RFF+BFN:S.KON.13'LIN+8'PIA+5+0198549512:IB'IMD+F+010+:::Evans'IMD+F+011+:::Julian'IMD+F+050+:::A wood of our own'IMD+F+110+:::Oxford'IMD+F+120+:::Oxford University Press'IMD+F+170+:::1995'IMD+F+180+:::158p'IMD+F+181+:::?: ill ; 25cm'IMD+F+270+:::76020000?:Forestry'IMD+F+300+:::Note?:Julian Evans ; with illustrati:ions by John White and Stephen Evan'IMD+F+300+:::s'QTY+21:1'FTX+ACB+3++TEST EDILIBE'PRI+YYY:17.99:CA:SRP'RFF+QNB:00023302:4'DTM+171:19960208:102'RFF+LI:8225'RFF+BFN:S.KON.23'LIN+9'PIA+5+0333628896:IB'IMD+F+010+:::Barraclough'IMD+F+011+:::Solon L'IMD+F+050+:::Forests and livelihoods'IMD+F+060+:::the social dynamics of den in devel:oping countries'IMD+F+110+:::London'IMD+F+120+:::Macmillan Publishers'IMD+F+170+:::1995'IMD+F+180+:::259p'IMD+F+181+:::; 23cm. - Bibl.?: p.245-253. - Inde'IMD+F+270+:::76302500?:Tropical forests'IMD+F+300+:::Note?:by Solon L. Barraclough and Kr:rishna B. Ghimire'QTY+21:1'FTX+ACB+3++TEST EDILIBE'PRI+YYY:45:CA:SRP'RFF+QNB:00023302:6'DTM+171:19960208:102'RFF+LI:8227'RFF+BFN:S.KON.23'LIN+10'PIA+5+0333645545:IB'IMD+F+010+:::Tyrrell'IMD+F+011+:::A J'IMD+F+050+:::Eiffel object-oriented programming'IMD+F+110+:::London'IMD+F+120+:::Macmillan Publishers'IMD+F+170+:::1995'IMD+F+180+:::283p'IMD+F+181+:::; 24cm. - Bibl.?: p.281. - Index'IMD+F+190+:::Macmillan computer science series'IMD+F+270+:::61305015?:Object-oriented programmin:g'IMD+F+300+:::1st Series Id?:E1108662'QTY+21:1'FTX+ACB+3++TEST EDILIBE'PRI+YYY:16.99:CA:SRP'RFF+QNB:00023302:7'DTM+171:19960208:102'RFF+LI:8228'RFF+BFN:S.KON.13'UNS+S'CNT+2:10'UNT+165+0000090001'UNH+0000170001+ORDERS:D:93A:UN:EAN007'BGM+224+B00405'DTM+137:19960209:102'NAD+BY+++STADT- UND UNIVERSITAETSBIBLIOTHEK:FRANKFURT+BOCKENHEIMER LANDSTR. 134-138+FRANKFURT+++DE'RFF+API:DE114110388'RFF+IT:STUB'NAD+SU+++B.H.BLACKWELL'CUX+2:GBP:9'LIN+1'PIA+5+0471960047:IB'QTY+21:1'FTX+ACB+3++TEST EDILIBE'PRI+YYY:40:CA:SRP'RFF+QNB:00023302:12'DTM+171:19960208:102'RFF+LI:8219'RFF+BFN:S.KON.E.631'LIN+2'PIA+5+0126487200:IB'IMD+F+050+:::The Multimedia dictionary'IMD+F+110+:::London'IMD+F+120+:::Academic Press'IMD+F+170+:::1995'IMD+F+180+:::351p'IMD+F+181+:::; ; 23cm'IMD+F+270+:::61756500?:Multi-media applications'IMD+F+300+:::Note?:edited by H. Sleurink'QTY+21:1'FTX+ACB+3++TEST EDLIBE'PRI+YYY:19.95:CA:SRP'RFF+QNB:00023302:2'DTM+171:19960208:102'RFF+LI:8223'RFF+BFN:S.KON.13'LIN+3'PIA+5+0333556453:IB'IMD+F+010+:::Medema'IMD+F+011+:::Steven G'IMD+F+050+:::Ronald H. Coase'IMD+F+110+:::London'IMD+F+120+:::Macmillan Publishers'IMD+F+170+:::1994'IMD+F+180+:::205p'IMD+F+181+:::; 23cm. - Bibl.?: p.192-203. - Inde'IMD+F+270+:::27651202?:Economic analysis/theory'QTY+21:1'FTX+ACB+3++TEST EDLIBE'PRI+YYY:40:CA:SRP'RFF+QNB:00023302:5'DTM+171:19960208:102'RFF+LI:8226'RFF+BFN:S.KON.07'UNS+S'CNT+2:3'UNT+55+0000170001'UNZ+2+72'

Name: Anonymous 2007-10-30 8:58

XML is still very bad.

Name: Anonymous 2007-10-30 9:03

xml is bad because?

Name: Anonymous 2007-10-30 9:33

thats better than XML

Name: Anonymous 2007-10-30 10:54

>>3
Because?
1. Fucked up structure (#text, attributes vs content, order matters - sometimes, etc.)
2. Unnecessarily verbose <LOL> </LOL>
3. Stupid parsing rules around EOLs and such, unsafe
4. Stupid namespaces and enterprise features nobody needed if they used it for what it was supposed to be used for
5. Full of enterprise bullshit and enterprise morons abusing it everywhere, filling the world with their enterprise XML-based shit
6. Suboptimal; I wanted a portable, multi-language, binary-safe, flexible grammar processing tool that could export an object model based on *any* grammar I could define so I could use existing file formats, binary data, etc.

Name: Anonymous 2007-10-30 11:35

>>5
1. You can say that of any format. A lot of fucked up structure our there is due to an entrprise idiot making a bad schema.

2. Agreed, but that's the whole "Markup Language" thing there. You can do standalone tags like this <tag /> (e.g. <br /> in XHTML)

4 & 5. All systems get enterprise bullshit added if they're used by enterprise. Everything they touch turns to shit.

6. Agreed wholeheartedly. I doubt it will ever happen though every tool has something wrong with it.

Barebones XML can be useful. I like tinyxml's minimal implementation quite a bit. It's nice when I need to throw together a file format quickly and am too lazy or pressed for time to make my own parser.

The biggest problem is the XML fad where people try to shoehorn everything into XML when something else would be more appropriate.

Name: Anonymous 2007-10-30 13:38

s-expressions > xml
(lol ...) > <lol>...</lol>
lisp is just xml after all. xml ~= s-expressions.

sexprs just don't require gay verbose </CLOSETAG>

Name: Anonymous 2007-10-30 13:57

>>7
wrong

Name: Anonymous 2007-10-30 19:06

>>5
PROTIP:Everything in life has a history. Those that do not learn from history are doomed to repeat it.
XML has a function and a scope. Use it within the right scope.

Name: Anonymous 2007-10-30 19:22

XML is for people who are too lazy to design and document a good binary format.

Name: Anonymous 2007-10-30 19:26

>>10
good binary format.
Such as? (not trolling, XMl sucks balls and I would like to design and document a good binary format.)

Name: Anonymous 2007-10-30 19:29

guys,
why can't i make a new thread?
fuck you shiichan
the specified thread id does not exist

Name: Anonymous 2007-10-30 19:48

>>9
When is the last time anyone used XML in the right domain? I don't think it ever happened.

Name: Anonymous 2007-10-30 19:52

XML is good, you guys are idiots.

Enjoy your hackish crappy formats.

Name: Anonymous 2007-10-30 19:53

Name: Anonymous 2007-10-30 22:14

>>13
odf. it is not binary so it is platform independant. a clear, consistant way to markup everything in the document to make it immediately useful to any compliant program. it is possible to be human editable which is perfect for something as important as documents.

Name: Anonymous 2007-10-30 22:34

>>5
lol, you're a tool. none of your points make even the slightest sense.

Name: Anonymous 2007-10-30 22:54

>>16
it is not binary so it is platform independant
That makes no sense. Ignoring endianness issues, binary files are platform independent.

Binary files are also as clear and consistent as XML. It's totally unnecessary to make your files human readable unless you actually intend for humans to edit them (e.g. config files). How many people are going to manually edit a putatively editable ODF file? None. Even if they wanted to, they couldn't figure out how. For all that it's made of readable letters, it's harder to read than a binary file in a hex editor.

Name: Anonymous 2007-10-31 0:19

>>18
Any plain text file is automatically NOT platform independent, due to CR/LF and character encoding issues.

Agreed, binary FTW.

Name: Anonymous 2007-10-31 3:17

>>12
Race condition.

Name: Anonymous 2007-10-31 3:32

>>8
no u

Name: Anonymous 2007-10-31 4:08

>>18
Once upon a time, documents were platform dependant. A document written on one platform could not be read on a different one. HTML was born of the need for a platform independant way of hyperlinking documents. It was designed to be human readable/modifyable to further that ideal. XML was born from HTML as a way to describe the semantics of general data as opposed to describing the semantics of a hypertext document. XML is designed for HUMANS. It is not designed to be fast. It is not designed to be efficient.

In practise, I can speculate that 99.99999999% of the time, all data described in XML would be parsed/modified without any human intervention. XML exists and it fills a realistic purpose. But if you like binary, have you heard of EBML? http://ebml.sourceforge.net/

Name: Anonymous 2007-10-31 4:36

>>22
Weren't both XML and HTML born from SGML, which was designed to be an easy machine-readable way to store data for long periods of time? An attempt to sidestep the "what the fuck do I open this with" problem. HTML, as far as I can tell, was just intended to be the simplest human-legible markup possible, based on SGML for who the hell knows what reasons. When it didn't cut it in a becoming-web-twenny world, they backed up a step and created XML, a simplified SGML. The design goals were to look similar to HTML and be verbose. It's as bad as Sepples.

Once upon a time, documents were platform dependant. A document written on one platform could not be read on a different one.
When was this? That didn't happen. They were only platform dependent in the sense that their structure was ad-hoc and unpublished. You can do the same thing with XML. With the space XML wastes, we could distribute a human-readable description of our binary format with every file using it.

Even if the reasons behind XML are good, the final product is garbage. And they knew it was going to be. "Do it fast" was a design goal. They didn't bother to strive for human readability. Only human-legibility. Meaning I see letters and angle brackets, but they don't make any sense.

Name: Anonymous 2007-10-31 6:03

>>7
I wish this were true. However, there are some fucking disgusting issues that make XML much harder to redo with s-expressions if you want to keep all of it, such as the behaviour of free text and the retarded difference between element attributes and subelements.

>>13
Lol, truth

>>14
Enjoy your ENTERPRISE software solutions.

>>16
it is not binary so it is platform independant.
You fail on several levels. Also, ODF and OOXML are memory dumps with angle brackets.

>>17
gb2/your enterprise, your manager needs to increase his profits and wants you to reduce TCO, now GTFO my /prog.

>>18
This man speaks the truth.

>>22
In practise, I can speculate that 99.99999999% of the time, all data described in XML would be parsed/modified without any human intervention.
Which means it was stupid to use XML in the first place. But it's all the rage, it's enterprise, you gotta use it for your mission-critical business solutions.

Name: Anonymous 2007-10-31 7:00

>>1
We've already discussed this http://dis.4chan.org/read/prog/1141396783

Name: Anonymous 2007-10-31 7:13

Name: Anonymous 2007-10-31 9:14

>>11
PNG is very good.
SWF has acquired some cruft over time, but is still pretty neat.
Unfortunately, most of the other binary formats I'm familiar with aren't exactly spectacular.

Name: Anonymous 2007-10-31 10:35

>>24
Enjoy your ENTERPRISE software solutions.
Enjoy your Hello-World-Scheme-Pi-Factorials-Programs.

Name: Anonymous 2007-10-31 10:46

>>24
gb2/your enterprise, your manager needs to increase his profits and wants you to reduce TCO, now GTFO my /prog.
GB2 dreaming about ever working with something bigger than the outdated PC under your desk.

Name: Anonymous 2007-10-31 13:34

>>29
Why do you assume working on the biggest project ever is supposed to be your professional goal?

My professional goal is to make the money I need while I have the most possible fun. This means no enterprise. Currently I work at a small company, on 1-2 man projects that last 3-12 months. I like this job. I wouldn't trade it for a position in some bigshot suited bloated business software consulting company where I have to write Java or COBOL with a shitty framework in a shitty XBOX piece of bloat bound to fail, even if they paid me twice as much.

Name: Anonymous 2007-10-31 15:23

>>29
Have fun fellating business majors just so you can feel important because they buy you toys. Like a dog begging for treats.

Name: Anonymous 2007-10-31 16:03

Wan wan~

Name: Anonymous 2007-10-31 22:24

are business app boring?

Name: Anonymous 2007-10-31 23:27

XML isn't bad in and of itself -- it succeeds at being a markup language that is trivially machine readable and easy for humans to understand. Is it bloated? Sure. But so are lots of other markup languages.

XML sucks because ENTERPRISE programmers misuse it -- think about it.

Fuck, I've seen a dumbass programmer pass this shit over a socket (you think I am kidding BUT I AM NOT:

<sometoolname>
<status>1</status>
</sometoolname>

Jesus fucking Christ. I always ask them why they DO THAT, and the answer boils down to some dumbfuckery like "easy to detect errors", "understandable by machines and humans", etc. Then they show me, with pride, this fifty fucking line program that opens a socket, waits for input, and BUILDS A FUCKING DOM TREE TO GET THE 1 OUT.

Then I show them my program that I wrote in C that does the same thing three times as fast, with no unneeded dependencies (minus standard net libraries), and all I pass is ONE FUCKING BYTE ACROSS THE NETWORKS. PROTIP: Mine is much easier to parse, thank you.

I blame "Computer and Information Science" curriculum, which trains these idiots in all the latest rage (XML, whatever language Microsoft decided to add to .NET, etc) and puts them out in the world... they don't have a fucking clue what's really going on.

Name: Anonymous 2007-10-31 23:48

>>34 XML isn't bad in and of itself
Sure it is. It's a misguided attempt to solve a slight problem with the biggest steamroller we can find. And WTF is with attributes?

Name: Anonymous 2007-10-31 23:51

>>35
If you're thinking of XML as it applies to a given "slight problem" you aren't thinking about XML in and of itself. GTFO and learn to read.
There are right ways to use XML and wrong ways. The fact that enterprise faggots nearly always use it wrong says nothing about XML itself.

Name: Anonymous 2007-11-01 0:23

>>36
I'm thinking of XML as it applies to the problem of data interchange between programs, which is arguably its purpose. Coming up with something that works is pretty easy, which is why I call it a slight problem. Coming up with something as bad as XML takes work though.

Name: Anonymous 2007-11-01 0:28

>>37
why not just use inf file syntax then? thats what XML file is isnt it? like header->data->close->header... etc

why all the html bloated bullshit added, if we just adopt a already existing syntax things will be much easier

Name: Anonymous 2007-11-01 0:37

>>38
Well, that's pretty much what I'm saying. It doesn't even need to be plain text. As far as I can tell, XML's only good point is that it already exists and it looks kind of like HTML with tumors.

Name: Anonymous 2007-11-01 2:15

New Non Retarded XML v1.0

element(attribute=value;attribute=value){element-contents}...

THREAD OVER
XML OVER


Name: Anonymous 2007-11-01 3:07

>>40
LOL

CSSXML

Name: Anonymous 2007-11-01 4:33

>>34
I agree that passing this xml trash over socket is stupid, but it's better than your 1 byte. It may be ok in your 20 line program in c to send that, but for large project you need to write your protocol so that whoever receives that ``1'' would understand it's sometoolname's status and not some noise, or some other valueable information. If you don't want to write your protocol, you use xml (i'd use ebml though)

Name: Anonymous 2007-11-01 6:11

EBML looks like shit. Variable length integers? Dates in nanoseconds offset from 1st Jan 2001? What the fuck were they thinking??

Name: Anonymous 2007-11-01 6:41

Beats CSV files.

Name: Anonymous 2007-11-01 11:49

>>39
The plain text requirement exists only because programmers are no longer manly enough to understand binary file formats. Kinda sad, really.

Name: Anonymous 2007-11-01 11:57

>>45
I reckon it's more historical. Back in the days when data interchange needed to be between very different systems - some with non-ASCII character sets - it made sense to make things as plain text as possible. You can see that sort of thinking in old Internet protocols such as SMTP and HTTP.

Name: Anonymous 2007-11-01 12:00

JSON

Name: Anonymous 2007-11-01 14:59

>>40

Not a markup language. Definately good for what a lot of people try to do in XML. No good for what a markup language is good at though.

Name: Anonymous 2007-11-02 6:27

>>46
Don't be retarded.

Name: Anonymous 2007-11-02 6:43

SPOILERS:

MARKUP LANGUAGES != PROGRAMMING LANGUAGES

Fucking hell, the company I work for is being comissioned to create a safety training "game" for the national institute of health, yet they're using the Unity game engine (LOL MACS) and they use XML to script the dialog and quest content.

It pisses me off.

Name: Anonymous 2007-11-02 8:57

>>50
Wrong. It depends how the data is interpreted as to whether it is a programming language or not.

Name: Anonymous 2007-11-02 9:54

>>51
XML was designed as a markup language.

Trying to use it for programming is cruise control for fail - that doesn't mean it's impossible, it's just extremely fucking shitty.

Name: Anonymous 2007-11-02 11:04

<for><initial><vardecl><type>int</type><name>i</name><value><integer>0</integer></value></vardecl></initial>...

Name: Anonymous 2007-11-02 11:14

>>52
You fail it, >>53 is made of win and awesome.

Name: Anonymous 2007-11-02 11:29

>>52
<?xml version="1.0" ?>
<program language="C">
  /* LOL version 1.00 beta */

  #include &lt;stdio.h&gt;

  int main(int argc, char **argv)
  {
    printf(&quot;LOL&quot;);
    return 0;
  }
</program>

Name: Anonymous 2007-11-02 11:37

BBCode > XML; I don't see how XML improves over it's predecessor, BBCode, in the slightest.

Name: Anonymous 2007-11-02 11:40

fuck i fail it

Name: Anonymous 2007-11-02 16:14

PHP > Perl; I don't see how Perl improves over it's predecessor, PHP, in the slightest.

Name: Anonymous 2007-11-02 16:15

C++ > C; I don't see how C improves over it's predecessor, C++, in the slightest.

Name: Anonymous 2007-11-02 16:50

Haskell > Lisp; I don't see how Lisp improves over it's predecessor, Haskell, in the slightest.

Name: Anonymous 2007-11-02 16:51

>>58
>>59
This is no good. You fail it.

Name: Anonymous 2007-11-02 16:51

>>60
You too.

Name: Anonymous 2007-11-02 16:53

>>61
Fuck you
>>62
You too.

Name: sage 2007-11-02 16:54

>>61,62
OH YEAH, YOU THINK YOU ARE TOUGH HUH?
I BET YOU CAN'T EVEN TUNE A FISH.

Name: Anonymous 2007-11-02 17:37

>>55
#include <iostream>
using namespace st;
int main()
{
  cout<<"LOL";
  return 0;
}

Now, which one looks the most efficient?

Name: Anonymous 2007-11-02 17:38

>>65
fuck, std*

Name: Anonymous 2007-11-02 17:44

>>65,66
This is what is wrong with C++.

Name: Anonymous 2007-11-04 15:50

XSLT can be considered a programming language if javascript can be as well. 

I'd just like to say that if you dont like xml because of the "bloated" characters, theres a nice compression algorithm you can use specialised for it. 

In that case any additional bloat is the fault of the person who defined the document.

Name: Anonymous 2007-11-04 17:06

>>68
Both XSLT and Javascript are Turing complete, so yes, they're full programming languages.

Name: Anonymous 2007-11-04 19:05

``ECMAScript''

Name: Anonymous 2007-11-04 23:52

>>67

C++ is full of std's

Name: Anonymous 2007-11-05 4:53

I'M RITCHIE
SON OF A BITCH C++
C++ IS PIG
DO YOU WANT CLASSES?
DO YOU WANT TEMPLATES?
C++ IS PIG DISGUSTING
BJARNE STROUSTROUP IS A MURDERER
FUCKING OOP

Name: Anonymous 2007-11-05 11:22

>>72
fucking lol

Name: Anonymous 2007-11-06 5:06

>>19
Ever heard of whitespace cropping and <br> tags?

Oh, and character sets are a problem with binaries too. At least *ML lets you specify which one you're using.

Name: Anonymous 2007-11-06 5:59

>>72
I rofled at that.

Name: Anonymous 2007-11-06 10:23

>>71
I lol'd

>>72
I didn't lol

Don't change these.
Name: Email:
Entire Thread Thread List