Archived

This topic is now archived and is closed to further replies.

EvilCrap

html? storage?

Recommended Posts

does anyone know how html is stored on servers? i think it might be stored literally - but then isnt that really ineffecient? if html were stored in a compiled form, then like 500+% more (pure html) space would be available right? and download would be faster. if its not compiled, then why isnt it? internet browsers could be distributed with a compiler, so distribution isnt an issue. why does html even exist anymore? its completely crappy. not a single modern web page is created with pure html, it feels kind of like ? some sort of skin that needs to be shed ? i think html should be abbandoned and replaced like DOS - dont you? Edited by - evilcrap on December 15, 2001 11:25:31 PM

Share this post


Link to post
Share on other sites
The myopia of priviledge... There are many people in diverse parts of the world who still run crappy old machines incapable of high-end graphical effects, but who can still browse pure HTML pages because those pages are simple text. There are also the blind and otherwise impaired who need pages to be pure text so they can easily be scanned for content to display in large fonts or use a screenreader. While you might think that some form of compiled page is an advantage, it presents security risks (What''s to stop a page from containing malicious code? What''s to stop current Java/Flash/Shockwave/other binary web applications from harboring virii and so forth?)

Moving to a compiled binary web will mean an increase in the amount of processing necessary to verify the material - checksums, encryptions and decryptions, etc - which may end up slowing the web down rather than speeding it up.

I agree with you in the long run, though, that HTML is starting to show both its age and its design limitations. It''s been patched and hacked with technologies like CSS, DOM, XML, etc, but we''re gradually approaching the stage where content alone is not enough (like television and the movies and videogames have). Flashy effects and a high level of graphical interactivity and control will be the paradigms of tomorrow''s internet, and many are trying to position themselves as the premier facilitators and conduits for that.

I''ll stop here.

[ GDNet Start Here | GDNet FAQ | MS RTFM | STL | Google ]
Thanks to Kylotan for the idea!

Share this post


Link to post
Share on other sites
HTML is stored as plain text, as it should be and should always be.

Personally I dont like flashy webpages that are filled with images and animations, I go for the content (and a nice color scheme). I disagree that the Internet should be filled with compiled web pages because I think that that is over kill. Oluseyi also brings up the good point of security. Plain text doesn''t hurt, compiled apps can.

HTML still exists because it does what it was ment to, give structure to the content on the Internet. It does it very well too.

Invader X
Invader''s Realm

Share this post


Link to post
Share on other sites
Guest Anonymous Poster
And remember, sites like gamedev.net store their data in databases and have html only as empty templates where to insert article from db.

Share this post


Link to post
Share on other sites
quote:
Original post by EvilCrap
does anyone know how html is stored on servers?
i think it might be stored literally - but then isnt that really ineffecient?
if html were stored in a compiled form, then like 500+% more (pure html) space would be available right? and download would be faster.

Most data transfer is compressed anyway, especially over a modem. Text compresses quite nicely.

quote:
internet browsers could be distributed with a compiler, so distribution isnt an issue.

Browsers could be distributed with everything include, so we could download a 650MB browser package and need nothing more. Wait: that''s Windows 98 onwards, isn''t it.

quote:
why does html even exist anymore? its completely crappy.

Back up your statements. HTML is human readable, it''s easily compressible, it''s highly structured, it''s cheap to create and edit, it scales well, it allows separation of form and appearance (when used properly with CSS)... HTML is one of the better standards.

quote:
not a single modern web page is created with pure html

Totally incorrect. Many modern web pages are created with pure HTML. I know this from having worked in the industry. Even pages created by PHP or ASP or whatever spit out HTML at the end as a decent common standard (of which there must be one). Next?

quote:
i think html should be abbandoned and replaced like DOS - dont you?

No. See my above statements and those of pretty much everyone else on this thread.

[ MSVC Fixes | STL | SDL | Game AI | Sockets | C++ Faq Lite | Boost ]

Share this post


Link to post
Share on other sites
Guest Anonymous Poster
i dont care about the graphics, graphics are not html, they are graphics.

html is scripted ( and in english too), and designed to be. crappy, crappy.

they could object orient and stream line html significantly more, as a new standard.

html results in the ugliest, most tangled, code i have ever seen.

web pages could be written with an API, maybe like win32 api ir mfc.

systems and readibility are an issue?
if compilers (like scriptors) were contained withing browsers, then you could go to "view source", and read your heart out.
HOW WOULD THIS EFFECT LOW END SYSTEMS?!

Security is an issue?
how is that?!
you could have a compiled set of html and still script it...
it would just have slews of less chars to process.
scripting is perfectly utterly insanly safe.


Share this post


Link to post
Share on other sites
I think you forget the origional concept behind the web. A HYPER TEXT MARKUP LANGUAGE = HTML.

It''s concept was never designed with dynamic content, it was meant as a means for any normal person to learn a few tags and publish documents to the web.

It wasn''t until later that dynamic content began to complicate the web so that now it borders on programming.

I don''t think HTML needs to be revamped because I feel it fits its true purpose, as a markup language.

Also, if it were compiled it would be more difficult and time consuming to create dynamic content. In addition, if you were to compile it for distribution, then you''d have to compile it across many platforms: Macintosh, Unix, Windows, BeOS, etc.

I think HTML does it''s job well. It''s platform independent (in its concept), easily accessible to anyone, and easy to develop (well, used to be).

Look into Zend for PHP, they actually do optimization to PHP code. I''m not sure if they pre-compile it on the server or what not. Might be exactly what you''re talking about.

R.

Share this post


Link to post
Share on other sites
Oh, I see, you want the web to be Java?

Won't happen for the same reasons I posted above. Java has its place, so does HTML. I do agree that to make HTML into what it is today, they complete ruinied it as it was designed, to the point that it's cludgy, ugly and difficult to work with.

The point of HTML wasn't to be super efficient, but to let anyone, anywhere, put up a document to the web and link them to each other. It was meant so that even my computer illiterate faimly could make a web page. HTML, when used as it was intended, isn't so ugly. It's the ever-growing push to take a web page beyond the static that causes problems.

HTML doesn't need to be replaced. Java and the scripting languages more tools to access common controls on web pages to make development easier.

R.

Edited by - Rube on December 17, 2001 11:23:58 AM

Share this post


Link to post
Share on other sites
quote:
Original post by Anonymous Poster (Evil Crap)
i dont care about the graphics, graphics are not html, they are graphics.

Graphics are what take longest to load on a web page, next to applets. There''s actually a reason why Google''s home page is so bare...

quote:
html is scripted ( and in english too), and designed to be. crappy, crappy.

You display an amazing ignorance of the design point and implementation of HTML. The HyperText Markup Language was designed to allow easy publishing and correlating to other documents on the then fledgling World Wide Web. You can find more information on HTML as well as associated technologies (CSS, DOM XML, XHTML, etc) at the World Wide Web Consortium.

quote:
they could object orient and stream line html significantly more, as a new standard.

Requires a ton more bandwidth. Inefficient and stupid.

quote:
html results in the ugliest, most tangled, code i have ever seen.

Liar. You haven''t seen Perl anyway.

HTML is very logically laid out and rather elegantly structured, especially when compared to the MFC source code...

quote:
web pages could be written with an API, maybe like win32 api ir mfc.

Now I''m sure you''re nuts.

quote:
systems and readibility are an issue?
if compilers (like scriptors) were contained withing browsers, then you could go to "view source", and read your heart out.

So you''re saying that the disabled should have to hit "view source" to get to content, in effect making them second-class citizens of the web?

quote:
HOW WOULD THIS EFFECT LOW END SYSTEMS?!

More processing. More bandwidth. Higher system requirements. If you can''t see it, it''s because you don''t understand software development.

quote:
Security is an issue?
how is that?!

*sigh*

I have other things to do. Someone please educate our young friend.

[ GDNet Start Here | GDNet FAQ | MS RTFM | STL | Google ]
Thanks to Kylotan for the idea!

Share this post


Link to post
Share on other sites
Guest Anonymous Poster
I agree with Invader X

quote:
web pages could be written with an API, maybe like win32 api ir mfc.


Ugh! Bad examples - the win32 api is overly broad and mfc obfuscates the underlying win32 api calls.

html is supposed to be replaced with xhtml and xml - both still text based - and then there''s an entirely new approach that blends text, script and bytecode together - Curl is an example of that.

IMO - the end of a text driven format will signal the end of the www - the network will still exist, it just won''t be recognizable - just another tv channel - same old marketing hyperbole and bs...

Share this post


Link to post
Share on other sites
Guest Anonymous Poster
What a retard this guy is! HTML is a markup language that defines the logical structure of an document! It is not a programming language. I repeat, HTML defines the logical structure of an document.

HTML contains no means of transferring data or securing it. HTML itself is the subject of encryption and transfer, that is why there exists protocols such as sftp or https.

By asking these ridiculous things you represent to be totally ignorant of the web technologies or the ideas behind them, or you''re just an dimwit.

Share this post


Link to post
Share on other sites
Pure HTML is great in it''s original design. Well thought out, and it fits the task nicely. However over the years with all the browser proprietary features, add ons, and lack of standards support the process of HTML development has been made a lot messier and over complicated than it needs to be.

HTML was designed to structure content - NOT define the presentation as rigidly as it currently does.

Hopefully when XML becomes more widespread this can be more so the case. Perhaps the best design would be structured XML content, which is transformed via XSLT into a new richer formating and presentation markup language that is yet to be developed for browsers that are yet to be developed. Flash can already be used this way to some extent.

Of course being XML/XSLT, the same content could also be transformed into straight HTML for legacy browser support.

HTML does have some interesting hacks for controlling the output display how you want, but it severely lacks the layout tools many graphic designers are used to in print/other media.

I pretend to work, they pretend to pay me.

Share this post


Link to post
Share on other sites
quote:

they could object orient and stream line html



Object-orientated design is a useful in imperative languages. HTML is not an imperative language. You do not create logical code in HTML; there is no such thing as an algorithm in HTML. HTML is not a programming language, and shouldn''t be designed like one.

I''ve designed professional websites, and I admit that HTML can be frustrating to work with if you try to do complicated formatting. The reason for this is that such formatting is not what HTML was designed to do. HTML was designed as a universal format for formatted text. That''s basically it.

Nowadays, many web designers do want more power (Primarily because their clients want more "WOW"). I don''t think that an imperative language is the answer. I do think that the major browsers should adopt reliable pixel and percentage positioning controls (IE, I think, is rather reliable in this department), but that''s about it. If we could use expressions like "100%-82" for objects X and Y coordinates and the space in which those coordinates are applied (window space, table cell space, etc) then I''d be happy. That''s all I want.

Share this post


Link to post
Share on other sites
quote:
Original post by Anonymous Poster
html is scripted ( and in english too), and designed to be. crappy, crappy.


That''s the beauty of HTML, plain English commands. The Average Joe could at the very least guess what each element is supposed to accomplish.

quote:
Original post by Anonymous Poster
they could object orient and stream line html significantly more, as a new standard.


Object orient it?! Wha? Why would you do a thing like that? HTML is meant to describe they structure of data (not the aesthetics, that''s CSS''s job). There is absolutely no need to make it more object oriented than it already is because it can already describe content efficiently.

quote:
Original post by Anonymous Poster
html results in the ugliest, most tangled, code I have ever seen.


HTML code is probably some of the most easiest to follow and understand. If you cant follow a tree, then yes, I guess it would look tangled - but then again, wouldn''t everything else?

quote:
Original post by Anonymous Poster
web pages could be written with an API, maybe like win32 api ir mfc.



  
HTML_API_DOC page = new HTML_API_DOC;
if (!page->Initialize(LONG List_of_Attributes))
return false;

ELEMENT text = new ELEMENT;
if (!text->Initialize(MORE Unnecessary_Attributes))
return false;

text->Modify(STILL_MORE Crap);

page->InsertElement(text);

//... repeat until page is filled



More to your liking?

quote:
systems and readibility are an issue?
if compilers (like scriptors) were contained withing browsers, then you could go to "view source", and read your heart out.
HOW WOULD THIS EFFECT LOW END SYSTEMS?!

Security is an issue?
how is that?!
you could have a compiled set of html and still script it...
it would just have slews of less chars to process.
scripting is perfectly utterly insanly safe.



I believe this was covered


Invader X
Invader''s Realm

Share this post


Link to post
Share on other sites