Curl Instead of Java or JavaScript? 252
janpod66 writes: "Tim Berners-Lee is putting his weight behind a new programming language designed by David Kranz intended to replace existing client-side programming languages like Java and JavaScript, as well as HTML. You can find more information at InteractiveWeek. Dertouzos, head of MIT's Lab for Computer Science is also involved.
You can also find more information at the startup company's web site. They have programming manauls on their web site. It looks vaguely like a mix of Tcl, Lisp and C (lots of low-level type declarations possible). They also provide a brief rationale. Now, I'm the first to admit that HTML, XML, DOM, JavaScript, Java, and style sheets have become rather complex. Actually, Curl looks pretty nice and clean. But does it stand a chance? And is going with something new, untried like this better than going with mature, widely understood technology?"
By the way (Score:1)
Re:Compatibility! (Score:1)
Here's the real scoop (Score:1)
Everyone is missing the real reason behind Tim Berners-Lee's advocacy of Curl. He's an investor and on the board of Curl. Need to say more?
Re:Nope, no chance. (Score:1)
Just look at the Common LISP Object System (CLOS).
Re:ANYTHING!!! (Score:1)
Java also runs programs as applications, not merely applets for your browser.
Re:ANYTHING!!! (Score:1)
Try starting your next project like this:
#!/usr/bin/env python
And you too will know the secret.
A language and web tool share the same name.. GRRR (Score:1)
http://curl.haxx.se/
has the same name as this.... erms.. thing.. though I can't verify it.. I do belive the curl.haxx.se folks have been using the curl name longer then the language dudes..
Too bad the curl.haxx.se folks didn't trademark.. now I'm going to have constant confusion with the term Curl!
I realy wish the language dudes would change their name!
Curl name already taken (Score:1)
As an erstwhile contributor to the original Curl, I am a little annoyed they did not bother to check freshmeat.net first to see if the name they had chosen was already in common use.
Other than that, I do not see how this technology has any serious potential unless they plan to submit it as a fully documented, patent-free international standard capable of independent implementation.
Great! (Score:1)
Re:Sun's spoilijng the game (Score:1)
Provide a code sample, or shaddup...
Simple, elegant, doomed (Score:1)
And, lose the monitoring crud and silly EULA; anyone who thinks they can make money from a tollbooth on an empty road is nuts.
--
Re:Congestion (Score:1)
Too bad we don't know those languages either (Score:1)
CobolScript
Nice site. When you mouse over a nav link it disappears. And the date is displayed as
Friday, April 6, 101
I guess it *is* Joe COBOL programming it.
Pains me to see stuff like that; I work at The Company Formerly Known As MicroFocus. :-)
Re:Won't reach critical mass (Score:1)
That's why we should give up on html and java/javascript and return to a language that everyone already has on his computer: Basic. Thanks to the diligent efforts of Microsoft and other companies, who have brought Basic into the 21st century, Basic already enjoys a bigger userbase than any other language (except perhaps Fortran's). Because so many programmers grew up writing their first programs in Basic, there exists a fluent userbase already. Basic is easily extensible and rather object-oriented when you consider its vast legacy.
Rather than interesting this post should have been moderated as funny. While not wanting to start a flame war I have to say that Basic is one of the least expressive languages that I have ever worked with. On top of that the syntax is just horrible. Just my 2 though. I know that there are many skilled vb programmers out there. It's just that I have never found one. ;)
Re:Sure! (Score:1)
Actually, with the Scripting Host support built into Windows, you can run client-scripts in IE and ASP-scripts in IIS in any language supporting it. This includes MS' own VBScript and JScript (of course) plus Python and Activestate PerlScript, at least. But those aren't deployed much so people simply assume that scripts in web pages == J(ava)Script and ASP == VBScript.
Re:My proposal (Score:2)
It seems to me that forcing the nested-container format makes HTML less confusing: one need not memorize the strange exceptions where they don't need closing tags. And it makes it easy for tools to find where the document is malformed. It's a win for novices and coders alike.
Re:Congestion (Score:2)
The difference, of course, is that I can use Java and JavaScript for the low price of FREE, and support is included in most browsers without the hassle of a plug-in.
Curl, on the other hand, wants to charge by the character for commercial content, and your customers will be forced to install their goofy plug-in to boot. Oh, and they want to force you to learn another language as well, that makes sense.
Face it, the folks behind Curl are clearly insane. Their mousetrap might be better at killing mice, unfortunately it irradiates the entire neighborhood and makes your house unsuitable for human habiation for the next 1e237 years.
Re:Some more words... (Score:2)
The minimum payment for commercial Curl usage is $1000 per month and the maximum is $50000 per month based on the characters of Curl content sent. I don't know what kind of magic Curl claims to be doing. But the bits from your images and text have to get to the client somehow. My guess is that they aren't doing a better job than PNG, JPEG, and Apache with mod_gzip.
Re:Java, anyone? (Score:2)
We have all seen what has happened to the VC driven Dot Com's and their startup kin. They thought (mistakenly for the most part) that the only database that was likely to be "good enough" was Oracle, and so they paid through the nose for it.
This made Oracle happy (in the short run) because they made a great deal of profits, but it made these profits by and large from a group of customers that are now unlikely to be repeat customers. Meanwhile the wiser startups (the ones you haven't heard of because they didn't buy Super Bowl commercials and haven't been listed on fuckedcompany.com) decided that some other database was "good enough" and they might actually be around again to buy upgrades in the future. If their needs grow from their success they might even actually need Oracle the second time around (but they will be able to afford it now). The fact of the matter is that for some workloads there simply is no choice but Oracle. They sell the only tool that is "good enough."
I can guarantee you that in the long run Oracle will either earn the premium that their customers pay, or they will have to lower their price (or go out of business). Oracle has plenty of competition, and most workloads do not come anywhere close to requiring their database.
Re:Commentary (Score:2)
Well said. There is no way that Curl has any chance of being succesful given that it is competing with several well tested (and well-known) systems that do essentially the same thing for free.
Re:Java, anyone? (Score:2)
Flash looks nice and has been embraced by the web design community due to it allowing better graphics/layout than all those cranky HTML/Java solutions and good tools for non programmers beeing available.
But its download time is a bad joke. It has spawned a sub genre of "loading"-flics, not unsimiliar to the pre-film fill material of older cinema days.
Its seems to run reasonably well on its few supported plattforms. (Except for the Netscape 6 desaster)
Java could have been very promising on the graphical side (look for example over there at John Maeda's applets [maedastudio.com] ) but it is also a PITA regarding download times (SWING, Baby!).
Plus Java's mutation rate has degraded the easiness of embodying an applet with a simple APPLET tag into a nightmare of version mismatchs between applets and browsers/VMs and forces users into using a Sun tool to construct a proper sh*tload of JavaScript that figures on what browser it sits and pulls in a proper Java VM in case the browser is too old.
On the other hand who would not forgive Netscape to give up the internal VM and stick that cr*p into an external plugin, if the internal VM was guilty of a big lot of bugs. (Run once, debug anywhere)
And of course has anyone written a Java tool suited for those non programmers that web designers are yet?
Thus Flash is partly useful to the artists and Java partly useful to the hackers.
If curl is useful to both it will have a chance on the web. It seems to address graphical expressiveness (and more than managing a fixed rectangle in the browser), intelligent downloading strategies, obligations as programming language, obligations for non programmers (providing authoring tools).
Its lack of cross plattfrom right now however is very bad.
I also have no clue, if any open implementation will be possible.
Who is going to pay? (Score:2)
I am all for making money off internet based products, but there are some things that don't lend themselvs to be "for charge". What would have happened if CERN charged for every character served out to an HTTP Browser? Do you think the net would have taken off? Just think of the cost of relearning a new language, making sure all your clients are compatable, and then paying for thier licencing... Customers commit to an annual volume of between $12,000 and $600,000, payable in equal monthly installments, with a minimum of $1,000 per month.
I say kill it now quietly before clueless CIO's get suckered into paying for something that will go the way of the DODO in a couple months...
Replace Java *AND* JavaScript?? (Score:2)
You know -- things like shell scripting, TCL, Ada, and C.
--
First Java, then XUL, then .NET, then Reef... (Score:2)
Yesterday, Eazel just announced Reef [gnome.org], yet another attempt to do the same thing Microsoft announced with .NET which is similar to Dave Hyatt's XUL [mozilla.org] (+CSS+JS) for the Mozilla project which promised to do what Java was supposed to do.
All this so I can subscribe to my word processor on a monthly basis?
Kinda depressing.
Seriously, can't we all get together and decide on a single system without everyone going off doing their own thing?
Answer: No.
-------------------
Re:Java, anyone? (Score:2)
Except, of course, the specs are freely available for anyone to implement. How is that "locked up"?
The key points are:
Java and Javascript won because they had corporate backing, and good "marketing" because they were 'in' webpages. Oh, and because they contain good ideas, and the proponents of the languages can actually communicate with the rest of the world and don't simply assume that the language will psionically make people use it by its mere existance.
Re:Sure! (Score:2)
Nope, Scripting Host works client-side in IE as well. You just need to specify e.g. "text/python" as the language. For an example, see my multiplication table [toriver.nu] (in Norwegian, but it demonstrates the concept).
Re:Won't reach critical mass (Score:2)
C-X C-S
Re:Won't reach critical mass (Score:2)
CobolScript [cobolscript.com]
Re:Replace Java *AND* JavaScript??.. CURL (Score:2)
Personally I am looking for a server side language that is better than JSP and Servlets. Perl is okay, and mod_perl is better, but can get messy, Php is good, but does not cache like Story Server.
On another note this is exclusionaly technology in its current form. It does not work with Netscape 6, it does not work on Linux or Unix at least that is what there system requirements say, no not even Mac either. So they inventeda new plugin.. BFD there are lots of those, and unless it is an open standard that anyone can implement then it is useles to me and would deter me from visiting that site.
Hey M$ is even talking about making there C# an open standard. Of course they wont make it open enough it will still have COM/DCOM/ActiveX hooks.
Who would have thought that the WWW would have become the World Wide Divide? Maybe they should detect a browser and say, "your browser maker is to dumb to make a browser that works with the standards! Please go away. (lol)
I don't want a lot, I just want it all!
Flame away, I have a hose!
/. nick "QuickBasic" == Bill Gates??? (Score:2)
Fine for static web, what about dynamic? (Score:2)
As for the "people who find HTML confusing enough already", in my experience, they mostly use products that allow them to avoid dealing with HTML.
CGI (Score:2)
--
Client Side Scripting SUCKS (Score:2)
Browsers as they exist can barely (and sometimes not) handle the much easier task of rendering a page, and until one put a standards compliant page on the web AND HAVE IT APPEAR SIMILARLY on 99.5% of the browsers in use WE DON'T NEED NO STINKING CLIENT SIDE SCRIPTING!!!
MOVE 'ZIG'.
Mature? Widely Understood? (Score:2)
Re:Java == client side? (Score:2)
Lynx (Score:2)
Integrating Code & Data? (Score:2)
I dont understand the rationale behind integrating them - they are 2 very different things. Developers need to be able to change one without the other, hence the recent rise of the old MVC paradigm in the JSP world (mixing the Java code in the HTML page makes for a mess because designers have to work around it, and it gets hard for coders to manage.)
So why does Curl suddenly think it grand to re-mix all the code and data? Can anyone explain this to me?
Aside from this, I don't see any reason to code in Curl. There's lots of mindshare and knowledge out there in HTML + DOM + CSS + JavaScript. Sure, it can be a pain sometime, but there are a lot of people out there working to make it better. Why turn that distributed responsibility over to a company with a proprietary technology? I'd rather just improve upon the current open standards.
My impression... (Score:2)
My initial impression of the language itself after looking at the text formatting example is that it looks hard to parse. There's no clear distinction between control syntax and data. I think HTML/SGML/XML looks easier to parse (at least initially)
Re:Um, Tard? (Score:2)
Though, having spent all day dealing with javascript, I'd rather that I could someday, within my lifetime, code to a single standard.
Then all I have to do is fix shitty dreamweaver code.
Boss of nothin. Big deal.
Son, go get daddy's hard plastic eyes.
Re:Especially when rebol is around. (Score:2)
As for the others it's already been ported to 45 platforms I am sure if the there is any demand it will be ported. How hard would it be port it tru64 unix when it already runs on aix, irix, solaris etc.
It's kind of odd that there is no palm support however.
Show me a man... (Score:2)
-------
CAIMLAS
Re:Mature? Widely Understood? (Score:2)
LiveConnect Replacement/MultiThreaded JavaScript? (Score:2)
We're planning on coming up with an alternative, cause JavaVm is slow to load, hard to configure on the client.
Does anyone know of a JavaScript only implementation of RPC server calls that works asynchronously (multithreaded) on both NS and IE?
Even a "synchonized" block in Javascript would be helpful. (You could use the multiple frames of a browser and use the *browser* to make the calls as a POST request and let JavaScript wait for the server to give a response, but it's hard to synchronize the JavaScript to control the frames....
Second Rule of Program Design (Score:2)
shell: #
C:
C++:
asp: '
curl: ||
SQL: -- (some, not all?)
Javadoc:
lisp: ;
vim: "
Forth: \
These are just the ones I know...
TIBET (Score:2)
A post of a different flavor... (Score:2)
- Classes
- Class Members
- Strings
- Collections
- Hash tables
...
- Memory Management
...
Note that this documents follows the 200 page "beginners/basic features guide"...I can see it now: "our language is so technologically advanced that you can use STRINGS if you become an expert in it." and "yeah well sure perl has nice hash's that they tell you about in the first chapter of learning perl. You can read about it on page 99 of the advanced features guide!"
-Chris [nondot.org]
CURL (Score:2)
The last time that I heard of a "CURL" was when it was in reference to that POS system that Vignette hacked together, and when I was shipped off against my will to one of their training seminars. They used to refer to "Custom" URL, where you do some thing ugly like wired news: http://www.wired.com/news/culture/0,1284,42774,00. html the comma separated shit actually keys to a cached file in your database system...
BTW, the first "0," means absolutely nothing.
Re:Probably Not (Score:2)
But that is a server side technology, so you can change your site over to php and nobody will even know (except that is has some cool new features). To adopt a new client side standard (like Curl) you need support from a lot of people: the major browser makers (to add support in their browsers), and most users (to upgrade to that new browser).
Most companies will still want to support older browsers anyway, so doing Curl will just add another level of complexity to the mess.
I'm much more interested in server side tech like XML/XSL that lets me use logical formatting for the data, but to then render it out in standard html (or SWF, or WAP, or PDF). If I want to change the data, I change only the data. If I want to change the formatting, I change only the template. I've been doing this for a long time with my own perl/database kludges, but it would be nice to have a standard format for it.
Some reasons why Curl will never take off... (Score:2)
1) It's a damn browser plug-in.
2) Does this sound familiar?
Surge [Curl] is currently for Microsoft Windows only - Macintosh and Linux coming soon!
Translation: Windows development takes priority and we release for this platform first (versus other projects that *truly* support all platforms and develop their releases in tandem).
3) How much space?!
Total installation will be ~ 17Mb depending on system configuration.
People will rush out ASAP and download this you say?
Riiiiiiiiiiiiiiiiiiiiiiiggggggggggggghhhhhhhtt.
Congestion (Score:2)
But it's in PDF. (Score:2)
Re:Good Languages Die Hard (Score:2)
http://www.curl.com/html/technology/documentation. jsp
I guess they can't afford to pay per-character for the use of CURL on their pages... :)
Easy! (Score:2)
--
How lame!! (Score:2)
--
Good Languages Die Hard (Score:2)
http://www.curl.com/html/technology/documentation. jsp
Seriously though, any new language is good, despite what anybody says. A different way to attack a problem always brings insight into solving others.
rationale is flawed (Score:2)
IMHO the basic rationale they provide for eth existence of CURL is flawed, based on mistaken assumptions. From their site:
re-drawing the entire page is a red herring - the issue here is speed of response, which is ruled by three factors: 1. processing speed, 2. available system resources, and 3. web connection speed. Their assumption is that server-side processing is inherently slower than client-side, but that isnt true nowadays by default because server-side processing actually is more efficient than client side on all these fronts.
the trend of most content-heavy websites is towards heavy duty server hardware with efficient server-side engines like PhP or mod_perl. Compare that hardware and software environment to a typical client-side desktop which suffers from M$ bloat, multi-tasking, and limited hardware. Unless you have server-grade hardware and don't use any other programs while surfing, you're probably not going to be able to execute code a.out in less time than it would take the server to execute it AND deliver it to you.
speaking of delivery, consider that the trend is towards cable modems, ISDN, and DSL and away plain old dialup. True not everyone has fat pipes (I don't , myself) but why design a technlogy for the past?
The first rationale basically is saying, "if you have a really slow connection, super-expensive hardware, and don't multi-task when browsing, you can benefit!" - what subset of users fit this profile?
odd they are talking about cost of delivery, when you can deliver content for free using PhP and HTML and it costs you PER CHARACTER to deliver content via CURL :P
granted, server-side processing sends the Whole Enchilada. Your 5-line PHP file expands to 100 lines of HTML and jscript. so what? if your connection is fast you won't notice, and if your hardware isn't state of the art you will wait for local compile time anyway. Just like you do now for Java (which is why IMHO Java sucks for web but rules for everything else).
well, actually if you follow CSS you can separate content and layout without installing Yet Another Plugin and also minimize complex nested tabling. Assuming you use Opera or IE anyway :) To be fair they should compare CURL vs HTML + CSS, not CURL vs HTML alone, if they want to talk about layout affecting code design.
they should rethink these issues a bit moreRe:fairly comprehensive (Score:2)
Just change the object.
Rate me on Picture-rate.com [picture-rate.com]
fairly comprehensive (Score:2)
Currently, the scripting option is not supported
Some of the syntax is a little strange, but I am assuming that it is related to LISP. It appears to support object and object orientated coding paradigms, but also incorporates some cool ideas from LISP - you can reassign a function pointer at runtime, something you cannot do in Java.
I wonder how this will stack up against Micro$oft's C# and the
-----------------
You are completely wrong (Score:2)
Slashdot editors and Java (Score:2)
Probably everyone to read this either (1) already knows that or (3) could care less and wants to get back to trolling. Well, fine. Have fun. But just keep that in the back of your mind.
Re: (Score:2)
some thoughts. (Score:2)
first, what exactly does curl offer that Java doesn't? It is a programming language plus client side environment ( say java), it offers a rich GUI,( say java) etc.
Second, the idea is that the separation of interactive code and static content is a hurdle. I don't think so. This separation allows different professionals to deal with their aspect of the product. It may be that some better way to deal with this separation is needed, but having every web developer be in charge simultanuously of programming the interaction and formatting the content seems to me less than desired.
Third, I am not sure that the need to combine various technologies is a liability either. This is after all a continuation of the Unix philosophy. I suspect that sooner or later someone will find out that Curl lacks something and people will come up with ideas of how to glue Curl and _your_favorite_technology_here_.
I do agree that the HTML browser GUI is unacceptable. What could solve this problem better would be a browser capability of delegating widgets defined in HTML to a toolkit ( tk, Gtk, etc.)
Re:Easy! (Score:2)
Turn 'em both off!!
Uhm.... (Score:2)
Link on the Curl download page: "System Requirements"
Not really two concepts that go together are they?
Re:What does their website use? (Score:2)
Your idea of "old school" is just that - old school. Typical Slashdot reactionary-ism. Just because this site runs off the cgi-bin/ doesn't mean that's the best way to do it. I guarantee you BEA Weblogic running in a cluster using servlets would absolutely devastate the performance of this site on similar hardware. Java (on the server side) is here to stay, at least for a while. You guys won't acknowledge it, but that's okay... the world moves on without you.
And no one is talking about replacing HTML, either. I'm not even going to go into how silly that sounds.
Re:Second Rule of Program Design (Score:2)
- Pascal: { }
- Basic: 10 REM
I'm pretty sure there was one that used '%'. Can't remember wich one though. Must have been one of those mainframe jcl type batch languages. I have visions of all-caps daisy-wheel printouts with that one, but as I said can't place it right now...Having a new one in a new language is asinine. It's fairly easy to tell yacc (or bison) to use all of #, //, and /* */ (the most popular comment indicators). To use || (a well-known operator in other languages) is doubly asinine, since apparently boolean OR is something else in their language...
JdV!!
charge? (Score:2)
Java == client side? (Score:2)
I think the best way to tell the difference is the salary difference between JavaScript and Java programmers.
This does not sound like a Tim move (Score:2)
So the PR description of Tim and Michael 'throwing their weight behind' Curl strikes me as unlikely and out of character. If Tim was interested in money he could have an options package almost certain to be worth seven figures from any number of profitable Internets.
What would be more Tim's scene would be a conference on scripting languages with lots of alternative proposals.
It strikes me though that rather than just rehashing for the nth time specification of yet another ALGOL or LISP like language it would be good to see use been made of other ideas that are not currently mainstream.
C# and Java do no more than a necessary and useful cleanup of C. But neither fits my concept of an object oriented language. These days OO has come to mean 'support for inheritance and methods bound to data structures'. I remember when the core idea of OO was considered message passing.
Surely there is much that could be done with message passing, direct support for parallel processing concepts etc.? The Inmos Occam language had many features that would be exceptionally useful in the context of programming for Internet applications.
Nothing could be less interesting however than merely tweaking the syntax of C to make it less eggregious. Syntactic sugar is important but ultimately an editor or preprocessor can be programmed to do the same work. Anyone can remember using RATFOR? Or the days when C++ was a preprocessor?
Re:This does not sound like a Tim move (Score:2)
I did use the past tense.
If by 'data abstraction' you mean a vacuous phrase used by marketting types in place of thought then I would agree that that is what OO has become.
Re:Curl name already taken (Score:2)
Re:Here today, gone tomorrow. (Score:2)
compulsive obssessive (Score:2)
Simplicity, ease of use, is not enough to compete with the marketing resources of Sun, so its going to be a difficult obstacle to overcome.
Now should they actually go over Sun, they would also have to hope company's would be willing to switch over from JAVA to CURL, and it can be difficult to convince a company to switch technologies altogether. Not only that, but you also run into issues such as, just how great this is on x platform running x backended to x, and when I say x I don't mean X as in the desktop environment. It does not have much by way of experience.
I like it though, its pretty neat.
curl free [antioffline.com]
Re:Won't reach critical mass (Score:2)
I saw it years ago. (Score:2)
I kept watching for something interesting to happen, but the page disappeared. A while later (still years ago), a "coming soon" commercial page popped up.
I mailed the guy who wrote it, and he made it pretty clear that it was going proprietary all the way.
A real shame. It was a great idea that got hung up for years while they decided to develop in private, hidden away from all the potential volunteer contributors, then killed by a decision to try to sell a programming/document language like a pile of cabbages. They don't seem to have realized that there have been better hypertext languages than HTML since before HTML was made, it's that HTML was an open standard supported by free software that made it so important.
--
You can keep your kitchen sink! (Score:2)
First off I downloaded SurgeLab to check out some of the code and the development environment. The curl code itself is extremely verbose, (aka takes lots of typing to create simple things) and is not very intuitive.
Second, I think the whole idea of an "everything including the bathroom sink" technology like this will make the development of even semi-large applications difficult. I much prefer to be able to seperate data from logic and to have a language for each that is taylored to its specific purpose. The whole idea of trying to mix display elements with the programming logic that controls the display is crazy. Anybody that has worked on even a decent sized web-app knows this.
Third, the licensing policy for curl is obscene. Can you imagine a start-up company having to fork over $1000/month (minimum) at the very begining just to pay for usage? Curl is the anthesis of the open source movement.
I saw a couple people remark that it would great to have a new web technology that is not controlled by one of computer giants. That's crap! Everybody wants to stand behind some David that is waging war against a giant, but I have no reason to believe that the people behind Curl aren't trying to make a giant of themselves.
I'd rather keep my text & images in resource & XML files, the display formatting information in stylesheets and XSL files, and the code all by itself.
Re:ANYTHING!!! (Score:2)
Javascript: an object oriented language interpreted by most browsers. It allows for client side dynamic html.
Java: an object oriented language that is compiled into an Applet which is downloaded to your browser and executed on a virtual machine. Because it is precompiled into an Applet, it is much faster than Javascript (after it's downloaded).
In general JavaScript allows you to dynamically manipulate the HTML on a page.
Java Applets allows you do things outside the scope of HMTL (like image manipulation or word processing).
Javascript and Java came out at about the same time, but the two really had nothing to do with each other. In fact, JavaScript was originally called LiveScript, but was renamed to capitalize on the Java name.
Probably Not (Score:2)
Re:What does their website use? (Score:2)
A Curl version of the Web site is in the works. Until some people actually have the client, there is not much point.
It is true that JSP and Curl can be complementary technologies (server-side production of a client-side applet). But in this case, the parallel version of our site will be designed to illustrate the advantages of moving most of the computation to the client.
What we're really doing .... (Score:2)
Unfortunately, the motivation behind developing the Curl language has not been accurately portrayed in this discussion. This is understandable due to many factors, mainly that our demos may not represent our vision completely (they are admittedly applet oriented while Curl is much, much more than that) and our pricing page not conveying the fact that employing Curl powered Web pages/sites can significantly reduce current operational costs such as Web hosting and backend infrustructure charges.
I have been developing database driven HTML/JS/CSS/DHTML/... Web pages/sites for quite some time now and have always run into the same problems. Cross browser incompatibility problems (not even just btwn NS and IE, but btwn different *versions* of the same browser), difficulty with maintainence, debugging, stability, and extensibility, and limitations inherent to the languages/protocols themselves.
One of Curl's missions is to provide a way for Web programmers get away from this restrictive situation by placing the programming power back into the Web developer's hands. Since Curl is a full fledged OO programming language (including multiple inheritence) with native support for text formatting and scripting, Web developers can very rapidly produce Weby UIs with the programming power only available outside of Web UIs - like in a C/C++/Java application. Sure you can embed a Java applet in an HTML page but only through liveconnect can that applet talk w/ JS - and then only through DOM can JS modify the page dynamically - now you are back dealing with cross browser incompatability issues
With Curl, you can really start thinking about advanced DHTML-like things in a cross browser environment with a development environment previously only available outside the DHTML world.
Additionally, there is a SAX2 compliant XML parser in the Core of the Curl runtime. So, writing a Curl Web page which gets its data from an XML stream opened to a server side PHP script based upon a user's interaction w/ the page (like clicking a button, filling out a form, or mousing over some text) is possible without embedding Java in an HTML page or refreshing the entire HTML page by applying an XSLT to the XML response. XML has been searching for a way to be employed on the client side in a Web environment and Curl is, in my opinion, a great answer.
How about graphics - many of the graphics on the Web are curves, gradients, and so on - graphics which can be procedurally generated. Curl's 2D library allows Web developers to reduce the number of little gifs their pages contain (obviously many situtations like pictures of people don't apply here).
And then there is 3D. With native 3D support, a Scene can be placed anywhere in a Curl page. Since it is not a third party 3D Scene embedded in an HTML page, but rather a single semantic framework that contains 3D objects, everything can talk to everything else. Drop down lists, buttons, text, etc. can contain events which interact w/ the Scene and vice versa. There is no break btwn components on the page - everything is uniform, knowledgable about each other, and live.
So, sorry for the rambling, I know that was long winded, but I just wanted get out some thoughts as to why I love programming in Curl and do not want to go back to the old environment where I'm always concerned as to whether my tables will line up right in both IE and NS. Let alone the whole other host of incompatabilities.
Open to discussion, I love talking about Curl...
Their pricing system is way open to abuse (Score:2)
Ok, let's say that some website that I don't like uses CURL. What will stop me from hacking up something that will keep reloading a certain page and causing them to be charged by CURL corp hundreds (if not thousands) of time. Even better, what if a CURL employee visits a site with CURL, s/he would be generating revenue for CURL corp. It can get more interesting. What if CURL corp. hires people all over the world (where US law does not apply, and FBI can't investigate) just to visit their customers' web-pages tens of times a day. The victim would not even know!
Just my $.02
Re:Won't reach critical mass (Score:3)
That's why we should give up on html and java/javascript and return to a language that everyone already has on his computer: Basic. Etc, etc.
When they "brought Basic into the 21st century", Microsoft had to introduce so many new features as to make Basic just as complex to use as Java or C++, or any other modern language really (ok, maybe I shouldn't go that far, they do take away a lot of flexibility that you would get from C or C++ for the additional cost of a lot more knowledge of APIs). Just because you learned Basic back in junior high doesn't mean you would have the foggiest idea how to use it's current and upcoming incarnations. Hell, Basic is the first language I ever encountered too. But I didn't stick with it very long. Other languages were much more powerful and made more sense to use. While the intricacies of any given language can take a significant effort to master, the effort can be well worth it if the language offers enough benefit over whatever you are currently using.
I suppose what I'm trying to say is that it really doesn't matter what language programmers grew up using. If it was Basic, they've long since moved on to other languages (including VB which is much much different than the Basic that many of us started off with), so it wouldn't do all that much good to move back to developing in Basic at this point.
Comment removed (Score:3)
It will do well (Score:3)
At LEAST as well as Scriptics Tclets did. You see those all OVER the place...
"Beware by whom you are called sane."
Re:ANYTHING!!! (Score:3)
What are you talking about?
Java applets are portable in exactly the same way that Java applications are. Both rely on the presence of a Java Virtual Machine having been ported to the computer on which you want to run your applet|application. The only differences between applets and applications are:
1. Applets by default have fairly severe security restrictions (can't access local file system, can't make network connections to any machine other than the one they were loaded from). Signing an applet removes this restriction.
2. Applets run inside of a browser window (or an Applet Viewer) and can take advantage of some of the resources of the browser (easy to launch another browser window, for example).
-jon
Problem not the standards but the tools. (Score:3)
Secondly I think that Jane Average will be able to take advantage of CSS, XHTML and DOM, they just won't need to know they are doing so.
We are just getting the next generation browsers that support these standards properly. Next is the software used to create webpages for those who don't want to code by hand. Such software will take care of the hassles of these standards for the user and allow them to just build what they want.
I don't have to understand postscript to print my word document, they won't have to understand CSS/XHTML/DOM to publish their page to a browser.
Especially when rebol is around. (Score:3)
It too is commercial but it's much much cheaper.
It has a tiny download and very small code which runs faster. It runs on just about any platform you could think of.
Oh yea the obligatory link [rebol.com]
rant on, brother man (Score:3)
Down with crap-ass workarounds!
"Smear'd with gumms of glutenous heat, I touch..." - Comus, John Milton
Chicken and Egg problem? (Score:3)
Likewise, the browser companys aren't going to build in support for an un-used language, because it's a waste of money....
Open Source to the rescue?!?!?
Compatibility! (Score:3)
I hate that I have to trap in my code for different browsers and handle them all differently. In case no-one at Netscape or MS know - the browser SHOULD SUPPORT XXX language the SAME, STANDARD WAY everytime!!
-----
Code and data integration (Score:3)
Applications - Web or otherwise - consist of code and data. In the current architecture of Web applications, there must be a strict partition between code and data because HTML can only describe data and has no ability to compute. But there is no real difference between an application and an interactive document. An interactive document is just an application housing mostly static data (text and graphics) with very little code (the interactive part).
I couldn't disagree more with this theory. After years of web development, including for one of the highest volume dynamic sites in the world, I believe there should be a strict separation between data, formatting, and interactivity. Every place I've worked has eventually come to the same conclusions:
- content writers don't want to know layout
- layout designers don't want to know programming
- programmers don't want to do layout or writing
Of course these are generalizations, but keeping these things seperate (at least keeping programming separate from the other two) has proven to work for the better. It's easier to find people, too.It's kind of like suggesting that a novel include alternating languages from paragraph to paragraph. Few people would be able to enjoy it!
Re:Commentary (Score:3)
Jay
Brilliant idea (Score:3)
Let's see: it's free to use if you aren't selling something, but if you are, then you'll have to pay metered fees. Shouldn't cost companies too much though, since the number of users who will download, install (and <cough> debug) the proprietary client plug-in will be negligle, so I guess the risk is minimal.
Oops, did we just invest 27 man-months and half a million dollars deploying a great new e-Commerce site entirely written in this thing?! And if we didn't want to go immediately out of business, we should have spent over twice as much deploying a non-Curl version of the site as well, and supporting both -- completely curtailing all of the so-called benefits?
And who's that I hear mumbling something about "separation of content and logic?"
Yeah, sounds like a winner to me. It's gonna fly like a lead balloon... but then again, that's what Keith Moon said to Jimmy Page, isn't it?
Won't reach critical mass (Score:3)
You have to reach a certain critical mass before you can dominate a market. This is doubly true for communication applications, because how can you communicate with a fellow on the other end unless you both speak the same language? Since we gave up on the glorious and noble enterprise of Esperanto, we've conceded the human-language field, but we're still working on the computer-language ones.
How can they expect enough people to adopt this new language? If you're writing for Flash or Shockwave, then you're already leaving out a lot of your userbase. Even if you write standard html, you're leaving out lots of user userbase, because of browsers' bugs (though IE, I have to say, doesn't have a single bug with rendering standard html). You're in the business of delivering your product, information, and you have to do it in a form your clients can understand.
That's why we should give up on html and java/javascript and return to a language that everyone already has on his computer: Basic. Thanks to the diligent efforts of Microsoft and other companies, who have brought Basic into the 21st century, Basic already enjoys a bigger userbase than any other language (except perhaps Fortran's). Because so many programmers grew up writing their first programs in Basic, there exists a fluent userbase already. Basic is easily extensible and rather object-oriented when you consider its vast legacy.
We should stick with the languages we already know and know well. It's better to improve what we already have than to open ourselves up to new incompatibilities and build ourselves another penthouse suite on the Tower of Babel.
Re:Chicken and Egg problem? (Score:4)
Re:Java, anyone? (Score:5)
Exactly. This is nothing more than a Flash or Java Applet substitute. Unfortunately for the folks working on Curl they seem to have forgotten the most basic premise of computer economics.
Curl is competing with several entrenched technologies, and both Flash and Java Applets have progressed a great deal over the last couple of years. More importantly, both of these solutions are easier and less expensive to deploy than Curl. So even if Curl has serious cool points it doesn't stand a chance.
What's most amazing to me is that apparently these folks just don't see that. That absolutely boggles my mind. Surely they must realize that the last thing that the web needs is yet another plug-in. Especially a plug-in that requires you to pay by the character for commercial content. The folks at Curl must be targetting the demographic of billionaires who recently had a botched frontal lobotomy.
Commentary (Score:5)
Free for non-commercial use, pay whatever they say for commercial use
.NET)
Basically, in today's environment, this will make it hard to get developer support. Open source tools or at least reliably free for use (Java) are the systems that will get adopted (exception: MS
Custom client simplifies client-server information sharing, using SGML-like language
Even if orgs want this, they are more likely to just use custom java client and XML. I don't see how there will be any substantial web browser support for this, so it will be just another plugin.
I definately understand the complexity of creating web apps, and they need to be simpler to create. But we should create simple frameworks for existing technology, and improve those platforms. I guess they think this will be some kind of quantum leap, but we'll see.
I got yer Basic right here (Score:5)
20 GOTO 10
Curl == Spyware (Score:5)
Then wander over to http://www.curl.com/html/products/pricing.jsp and look at the fact that you have to commit to sending Curl a minimum of $1000/month (max of $50,000/month) to use Curl to deliver content. And the cost is based on how many characters you serve. Not, on how much revenue it generates.
This product looks more like misguided megalomania than like product that stands a chance of actually being used by anyone.
Technically, it acutally looks pretty good. But, the business model and the privacy policy are, well... They're insane.
StoneWolf
Some more words... (Score:5)
Curl was dead before the press release.
Move along folks. Nothing to see here.
-----------------------------
kaaaameeeeeeehaaaaaameeeeeha!
-----------------------------