CURL -- similar idea to REBOL?

I saw CURL more as `display technology' than as `processing technology', but that was perhaps out of naivite, certainly _not_ from profound analysis. I find it very impressive as display technology, and would also be interested in hearing from anyone who has had substantial experience.

-- Ryan Cole Programmer Analyst www.iesco-dms.com 707-468-5400

I could see them having a high burn rate especially since recently they had received 40m in VC. Which almost seems unprecidented for this type of company. Its quite mind boggling actually.

-- Ryan Cole Programmer Analyst www.iesco-dms.com 707-468-5400

NQL www.nqli.com The thread was mainly about their spamming the list rather than their technology. Though I seem to remember someone quoting CS as saying he could implement NQL as a Rebol dialect. --Colin.

This is what I find so interesting. Already, I've found three paradigms (I call them "paradigms" instead of "languages" because in many ways they are much more than that) which attempt to answer the present lack of sophistication in today's solutions for network connectivity: NQL, REBOL and CURL. It seems they are also pushing client-side interactivity, though NQL seems like it does less of this (and I suppose CURL has less emphasis on network connectivity). I think these are the next big trends in software development: easy networking and richer client-side functionality. These three companies are probably just the tip of the iceberg. Prepare yourselves for standards wars for years to come. May the best technology win! Carl Youngblood

It's not really unprecedented to bring in > $40M for a company shipping enterprise applications. Groove (~$50M) and Zaplet (~$50M) spring to mind. Launching a platform that you want IT managers to bet their life on requires a huge credibility and long track record. Both of the above examples (and Curl) will be convincing big reference accounts to take a risk with them, implementing consulting projects below or close to cost. I'm betting that REBOL is doing exactly this with the unnamed silicon valley chip plant. So the big upfront infusions ensure that they'll be around after they've burned through their capital establishing credibility and platform maturity. I would attribute the stop and go /Express development (for example) to interruptions by paying clients. That's how it went at a previous employer of mine -> using consulting revenue to drive the software dev. process is pretty much the only way to avoid VCs, but it sure goes slow. Not taking on VC money definitely has advantages, but at some point you have to commit to becoming big enough for the conservative IT herd to bet on. JS

I looked them up and found this. Looks like a cash-flow problem... SANTA ANA, Calif., July 13, 2001 - NQL Inc. (Nasdaq NM: NQLI) announced today that the company is vigorously pursuing the sale of its NQL technology. According to NQL chairman, CEO and president Douglas J. Tullio, management has identified over 100 potential acquirors with the resources and access to markets needed to developed the technology to its highest potential. NQL will pursue the incorporation of a royalty arrangement with prospective acquirors in an effort to reward stockholders in the future for their investment in developing the award-winning NQL technology. The condition of today's capital markets makes it difficult to raise working capital on terms beneficial to NQL stockholders, commented Tullio. For that reason, we believe it is in the best interest of our stockholders to sell our software technology ......"

Watch out here, though. There are several projects called Curl. Each one competes with REBOL in some way, often not very well. The Curl mentioned on Slashdot recently is not free - only the client is free. The server is paid for by the applet download. This means that more traffic on your site means more money you have to pay to the Curl company. This makes Curl more expensive than Express in the long run. Also, in order to keep track of your traffic so that they can bill you for it, all Curl network traffic must refer to a central server; better hope that server doesn't crash, or so will yours. By the way, there are architectural limitations, such as the client being embedded in a browser, that would make it impossible to implement _this_ Curl in Perl. I think you may be talking about one of the other Curls, rather than the one Mr. Youngblood was asking about. The Curl you described sounds interesting, though. Do you have a link I could check? Brian Hawley

What you say is probably true, but I sure have trouble, reading the financial press, seeing how there's any `there' there for companies like Zaplet and NQL (now trading under 0.50/share, i.e. losing 90%+ of its value over the past year) I can't find much about `Groove' (unless it's `Modern Groove Entertainment Inc) though, so I can't comment on them. ...

Or, alternatively, perhaps to occasional rushes of sanity in an otherwise foggy futuristic cloud (of blue smoke?) ... After all, VC money is _very_ easy to find these days. It's just not in the hands of the VCs any longer....

Have you checked out their business model? If they actually get away with it, it's a money machine. They charge by the traffic, by _your_ traffic. All they have to do is get a high-traffic web site as a customer, just one, and they'll make back that money easily. Brian Hawley Silly question: Is your email name related to the band?

Well, let's see... Language Syntax: Vaguely object-functional. Ugly, but not as much so as the C family or (cringe) Perl. Reminds me of POV-RAY. Language Model: Somewhere between VID and NewtonScript, with some POV thrown in for good measure. Specifications are written in a style similar to VID. The object model is similar to NewtonScript, dual inheritance and such. 3D stuff is similar to POV-RAY. Overall, fewer capabilities than any one of these (let alone all three). The feel of the language is awkward, almost as if it were implemented by Perl programmers. The FAQs on the site mostly deal with understanding language weirdness and work-arounds for what I would consider design flaws in the language. Security model: Similar to the Java or REBOL/View sandbox model, but more strict. Worth looking at, especially for a /View browser plugin. Client-side Architecture: Client-side browser plugin that "compiles" Curl source or byte code into "display" code, a combination of HTML, JavaScript, Java, Flash and DirectX code, which means that you need all of these on the client side to use Curl content, many megabytes. Server-side architecture: Curl content could be served up by a standard web server, but since you pay for it by the download some way to keep track of downloads and sending that data to Curl servers is required. How this is accomplished is unclear on their site. Perhaps Curl.com serves up the content, or perhaps there is a module in your web server that sends them the data, or perhaps the browser plugin does it. Unknown. Developer tools: Because of the awkwardness of the language, it is good to have tools to generate the code without you having to see it. They have many such tools. As it is, I wouldn't suggest using Curl on your site unless you are expecting your site traffic to generate enough cash to cover the per-download charges. If you want accessibility features on your site, forget it (and REBOL/View too). The language could be implemented easily as a REBOL dialect though the 3D facilities would be tough without the native support. A REBOL dialect would be easier to use as well and less ugly. Overall, Curl would be a good thing to look at for ideas on the REBOL browser plugin, particularly its security model, development tools and seamless browser integration. REBOL should be able to beat this one easily :) Brian Hawley

I hadn't seen or payed attention to their business model or pricing scheme when I was playing around with the technology (I AM GUILTY OF NOT READING A LICENSE AGREEMENT *SIGH*) but yes, I see IF they could get it to work it would be a like a money tree. However I don't see them as even being at the stage where they would be able to command that sort of VC or Angel investment. Maybe down the road after the technology had been solidified a bit but it still looks like a infant language to me.

Silly Answer: yes

But VC's usually give money on the strengh/pedigree of the management team. The VC game is mostly about business not tech. dv

-- Ryan Cole Programmer Analyst www.iesco-dms.com 707-468-5400

Actually, no surprise there. Most great software was created by small teams, especially in the beginning. However, I have met some of the RT guys and they are not human :) -- in terms of their coding prowess that is.

of

The language has been around for a few years, I think I've still got a zip of the old MIT curl project that the commercial enterprise stemmed from. What surprised me during the Curl beta was that after a year and 250 people working on it, it showed surprisingly little advancement, especially considering what 4 guys at RT did in the same time. Cheers, Allen K

-- Ryan Cole Programmer Analyst www.iesco-dms.com 707-468-5400