Cisco To Open-Source New Messaging Protocol

Esther Schindler writes "Do you use SOAP, CORBA or EJBs? You might want to take a look at Etch, writes James Turner for CIO.com. It's language-, platform- and transport-agnostic, and Cisco is planning to release it as open source. Certainly, it offers some technical benefits: 'In addition to a simplified configuration, Etch also promises less overhead over the wire, compared to SOAP. In a testbed environment where SOAP was managing around 900 calls a second, Etch generated more than 50,000 messages in a one-way mode, and 15,000 transactions with a full round-trip, company officials stated.' And the open source part? Cisco is in the process of deciding what license to use. 'The intent is to use a less restrictive license than GPL, perhaps Apache or Mozilla. This is to allow commercial developers to incorporate Etch into products without licensing issues. A final announcement on the licensing decision will be available in the next month.'"

Comments

GPL

[Anonymous Coward]

Glad to see more and more companies moving away from GPL, understanding that it will only limit the potential adoption. As a highly respected registered member of the Slashdot community, I'm posting as AC as this post will very likely be modded troll.

Re:GPL

[Shadow-isoHunt]

You care too much about your karma - regardless of if your post is being sarcastic or not - say what you mean and mean what you say, stand behind it because we won't believe an AC anyways.

Re:GPL

[whatever3003]

Coward. Your priorities show a distinct lack of strength in your convictions.

Re:GPL

[superskippy]

Sorry to feed the troll, but the point of the GPL is not to increase adoption. Your absolutely right to say that other licenses will lead to greater adoption- but this is adoption by people who may take, take, take and not give back.
Besides, it sounds like LGPL is what's needed in this case, anyhow.

Re:GPL

[unix_core]

If it's not going to be kept free, then what significant benefits are there of it being adopted?

Re:GPL

[Teckla]

Sorry to feed the troll, but the point of the GPL is not to increase adoption. Your absolutely right to say that other licenses will lead to greater adoption- but this is adoption by people who may take, take, take and not give back.

The company I work for sells closed source software. We also use some open source software (not GPL) in the product.

We contribute back to the open source we use because it's more sensible. Adding the same features back in again and again would be counterproductive. We'd rather they get added to the open source project permanently.

We have a blanket ban on using GPL'd source, though. We can't afford to GPL our entire 20 million line software stack, which would be the result of using even a tiny bit of GPL code.

Try to understand that not everyone loves the GPL and not everyone that doesn't love the GPL is a troll.

Now it's my turn to get modded into oblivion for not being fond of the GPL. Sigh.

Re:GPL

[Kjella]

We contribute back to the open source we use because it's more sensible. Well, the positive side is that you're contributing something. But I'd rather say that as "We contribute back to the open source we use only when it's more sensible". That means trivial fixes, basic features and other things that doesn't threaten your business and is cheaper to "outsource" the maintenance on. Any time it's major features, more specific layers to the business you're in that could make it easier to produce a software stack like yours, most decide to pile it up on their 20 MLOC proprietary pile because the cost/benefit ratio swings in favor of keeping it in-house.

Taking a huge proprietary application and adding a tiny bit of GPL is foolish, the GPL is best designed to work organically - it almost does what you need, you add the missing pieces and you have to distribute the changes with your binaries. Both suffer a bit from the "big change" problem, there's rarely anyone with incentive to do major changes, but the GPL suffers less because companies are often "forced" to release code they wouldn't have released if they didn't need to. In practise, I find that means there's a lot more GPL code I want to use - actual end-user applications that otherwise would have gone into the proprietary stack.

Try to understand that not everyone loves the GPL and not everyone that doesn't love the GPL is a troll. As a producer of proprietary software, it's no trouble understanding where your preferances lie. You get to keep all the good and important parts to yourself, you get a free toolbox and you get some others to do free (as in beer) maintenance. What's in it for me? Unless I start using anything pure BSD with source, little if anything. Maybe a bit better/cheaper software but usually the proprietary derivates will only use that to increase their margins. Does OS X being based on *BSD bring the BSDs much good, or is it just a bleak shadow with sub-percent market share? Has it reduced the prices of OS X?

Stating your preference for BSD because it helps proprietary development is an honest opinion. But when you take on the role of end-user, the GPL means I can use it any way I want, I have the source to modify it any way I want and as a company you can do as much proprietary in-house modifications as you want. The only thing you can't do is sell/distribute proprietary versions to *drumroll* end-users. Every OSS developer including Linus himself is probably an end-user to 100x as many projects as they develop, so we're all end-users. So when someone tries to say to end users "You shouldn't use GPL software because it's not really free." they're have either completely misunderstood, are trolling or is pushing someone else's agenda. It's extra freedom, but not extra freedom that'll do the end users any good.

Re:GPL

[daffmeister]

That's because you've only met people similar to you. It's pretty normal to mostly encounter others of a like nature.

However, there are plenty of groups out there who would quite happily take GPL code and add it to a closed source app, if the licence allowed them (and some that will do so even against the licence). Just because you haven't personally met them, doesn't mean they don't exist.

Re:GPL

[Cyclops]

OpenBSD doesn't want to take GPL'ed code. They can, but they don't want to.

They are perfectly fine to include it (and they do include GCC, for instance), and even link to it (but then the derived work will have to be GPL'ed and they don't want that).

And some projects have the problem in the inverse direction. Linux can't benefit from dtrace except in design principles.

Or even the fantastic ZFS.

Oh well, I guess it's all down to the same premise: if you don't want/can't use it, then stop bitching and go write your own, spoiled brats...

Re:GPL

[Cyclops]

Except that giving a "BSD exception" to the GPL would make the point of the GPL kind of moot.

The GNU GPL is there to make sure every single user of GPL'ed code has the 4 software freedoms.

The BSD's only make sure for the first recipient.

I'm not claiming the first is better than the second (although I believe so), just that their purposes are different enough to make it a childish request instead of coding your own version like a real man.

Re:GPL

[nahdude812]

Under such a scenario, when the majority of source files in a BSD project contain a snippet of GPL code, and that project can no longer build without those GPL snippets, then the project is effectively GPL since the GPL code has become inseparable from the BSD code.

Or else the exemption allows the code to become relicensed as BSD when included in a BSD project - then I can start a "GNU/Linux/BSD" project which is simply a repackaging of regular GNU/Linux under a BSD license.

It's not the place of the GPL to provide such an exception, it's the place of project owners to provide such an exception for their project if they desire it.

FWIW, there's some awesome stuff in BSD, but personally I feel like if I'm providing my code to you, and you find it useful the least you can do is provide your changes to my code back to me so the rest of the world can benefit.

Re:Um, what?

[RodgerDodger]

You open-source a protocol by providing a specification with no attached IP rights, such as patents covering the protocol. A reference implementation kind of helps, too.

Re:Um, what?

[Timothy Brownawell]

Wouldn't that be "open spec" instead of "open source", with the open source reference implementation being a separate issue?

Re:Cisco, Please use the LGPLv3 license.

[Richard_at_work]

Sod the libraries, license them however you wish - give us full and unfettered access to the specifications so those of us that wish to produce a BSD licensed or Public Domain set of libraries can do so. Don't assume that any license you choose for the libraries today will be good enough for everyone tomorrow.

performance stats are probably misleading

[Anonymous Coward]

I bet the test transactions were trivial, so the performance numbers would be dominated by the speed of parsing, validating, and dispatching the message.

Not saying that SOAP is the best solution. It is XML-based, which everyone realizes is a mixed bag. In particular, validating XML parsers have to be huge to cope with the specification bloat. But why should everyone rush to accept such a fundamental infrastructure piece from a single vendor? Any messaging scheme based on the old TLV (tag, length, value) scheme stored in network byte order would beat SOAP in performance tests. And as features are added to answer the diverse needs of the various communities, I expect performance degradation in the whatever-comes-after-SOAP, too.

Re:Microsoft's Kerberos

[Richard_at_work]

So? That's called 'Freedom'. I'd rather have too much of it than too little.

Re:Less overhead, more messages - but ..

[K. S. Kyosuke]

Might be, but you are totally off-topic here. This article is about doing this ober network, which has nothing to do with Instant Messaging, save for the fact that some information transfer is involved. IIRC, there are things like XEP-0072, that allow for application data messaging over XMPP, but these are also not exactly of interest to most Jabber users, and if they require server extensions, well, you'll be out of luck with using "popular services" }the public ones?] to transfer such data.

Re:Distributed computing.

[gbjbaanb]

Both those protocols suffer from 1 problem: bloat. The reason they're bloated and inefficient is because a committee decided how and what to add to the protocol once it was initiated, and we all know how well that works out.

SOAP was a 'quick and dirty solution (by Don Box IIRC) to (apart from getting a job at MS :) ) transfer COM calls over a http tunnel instead of the usual DCE-RPC tunnel, and it worked well when you only wanted to send a request to an object. Obviously, it has to have a webserver on the other end which slows it down tremendously, and then they added support for all kinds of complex types and a large schema as well. I'm surprised it works at all after seeing the raw WSDL code!

CORBA... designed by committee to do everything including transport kitchen sinks.

Since I've been working in the industry there is a tendency for supposedly bright people to take something simple and 'make it a general purpose solution' or 'implement some framework features' which nearly always breaks it into a bloated POS far removed from the original, simple, easy to use, and effective solution.

I welcome Cisco's new protocol, I don't care if it doesn't do everything I might possibly ever want to do, as long as it does the majority of my work quickly and simply. I can work around the edge cases myself, possibly even (gosh!) redesigning the way those edge cases work.

Re:ZeroC's ICE

[eric2hill]

Their standard quoted price is $10K for unlimited royalty-free distribution, but they are *VERY* willing to work with you to price the product correctly for your product. Don't discount that number if you have a commercial application. Negotiating a percentage of sales opposed to writing your own communications subsystem is really a no-brainer.

Re:Um, what?

[Anonymous Coward]

Not all open specifications are freely licensed. The MPEG specifications are open, for instance, but you need to pay a license fee to use them.

I think we really need to come up with a better term for this, or narrow the definition of an open specification.