Slashdot Mirror
Novell to Ship MySQL With NetWare 6
An anonymous reader writes "Coming close on the heels of their announcement that they've ported PostgreSQL to NetWare, Novell announced today that they will begin shipping MySQL with NetWare 6. Owing to customer and partner doubts about the GPL, Novell has chosen the commercial version of MySQL, rather than the GPL'ed version."
Reading the license term for MySQL it seems pretty clear that Novell has no choice in choosing the license model. According to the terms at the MySQL page, MySQL is only GPL if the whole system is open sourced or GPL:ed. N'est ce pas?
It says it'll ship in 2003, but a beta is available.
WHEN will MySQL 4 get out of 'development' and into 'stable'? The infoworld article was already mentioning MySQL 5, but 4 is still alpha/beta, not 'production', and the 3.23 series seems to be progressing still.
creation science book
I loved Netware 4.11 and think that NDS even then is better than the next two iterations of Active Directory could hope to be.
However, Novell has been doing this "Me too!!!!" thing with bundling stuff for years. Perl, the whole Netscape server, some IBM web thing, etc and it means nothing.
I hate to agree with the trolls, but Novell is dying. There was even an article in the WSJ last Friday about companies trading *below* their hard asset valus, and guess who was on it? Novell was! The Wall Street logic apparently was that trading below asset value was the sign that you were a dead duck and that investors not only didn't think you would do well now, but thought you'd likely go bankrupt, too.
I dislike marketers as much as any programmer, or for that matter, anyone who's had their dinner interrupted by a phone call. But Novell needs better marketing in general. They've got really neat technology, but nobody knows about it. I think that if Novell discovered cold fusion they wouldn't tell anyone. The most they'd do is but a little paper sign on the door of the laboratory saying "cold fusion inside - don't tell anyone."
So, without the proper marketing, I doubt anyone will ever discover that Novell can be a web services platform, or that there's a built-in database that's ready to use.
Get your stinking paws off me you damn dirty ape
This is not really a big issue one way or the other for the GPL. MySQL is available with a GPL license, for use in GPL applications, or with a different license for non-GPL applications. Novell thought that some of their customers might want to build a non-GPL applications with MySQL and Netware, so they gave them that option by supplying the non-GPL license for MySQL. I mean, this makes sense; I would think that adding MySQL with the GPL license to Netware isn't much of a value add.
I were Novell, I'd be more interested in developing a Samba-style SMB server NLM to try to replace NT file and print servers
Netware 6 already has this. I forget their name for it but, it goes under the guise of Any Client or some such. With this feature, Microsoft clients can connect to the Netware 6 servers without the previously required Novell client. The Netware 6 server looks to the client, like a NT server. Netware 6 also supports an NFS like export that allows *nix clients to also connect natively, without the use of Novell client software.
Netware 6 has a lot of really powerful features. What's more, I think that Linuxers would like it because it has a similar feel, even if the commands are different. Hell, it even runs Xwindows with the IceWM.
(for the humor impaired, it's just a joke)
As much as I used to agree with those who sound the "better product" drum, as a former Novell sysadmin (primary NDS admin for a state university and developer of a YES approved NLM), Novell has lost it. They have too much development in Bangalore (yes, I have participated in a conference call with Bangalore engineers, and yes, they did speak English well, but didn't quite get some concepts on failover I was trying to explain as required) and too much turnover among developer staff. Couple that with a core kernel that's too small to extend (flaws in the kernel prevented effective multi-CPU work are documented on their developer site -- look for NKS) and you have a lean, mean server OS that rocks on a 486, but looks as out of place today as big hair and belts over large sweaters.
For example, Win2K was released with 100,000 known bugs. Apache Software Foundation was running their website w/Apache 2.0 beta for over a year before the code went "gold". This is the fundamental difference. Just b/c Microsoft calls it SQL Server 2000 doesn't mean it's gold code.
For years, Novell Netware included a copy of Oracle. Didn't Oracle recently announce that they aren't porting to Netware anymore? That's probably the only reason that Novell has moved to support MySQL and PostgreSQL.
Slashdot: come for the pedantry, stay for the condescension.
Yet if you look at their full product range they have products such as;
I could go on but the message is clear, the company is packed with good products which it doesn't know how to sell.
Last year I ran an evaluation of all the Meta Directory software out there and DirXML was the clear winner. We bought it and are very happy with it's performance, it certainly should be looked at by anyone who has looked at the Sun ONE or Siemens "equivelents".
My advice to Novell would be that they need to spin off the Netware business to continue developing this and keeping their many millions of existing users happy. The remainder of the business should then be refocused as a Directory Services company. They already almost give away eDirectory, they should make this more official and then when organisations are hooked sell them all the value add products which integrate so nicely with this.
This would also be welcomed by all the organisations who are concerned about Active Directory's single platform nature and the high cost of the Sun ONE Directory and their on|off support for Linux, which Novell have always been very committed too.
And if they're not 100% sure, why are they touting it? For the same reason that commercial companies release before it's time: mindshare.
Yes, we know that an open beta from Apache is as good or better than an initial release from a commercial developer. We know that the stable releases from Apache are akin to the third (or so) patch release for a commercial product.
What's the difference? Quality? Not necessarily. The difference is in semantics. Some open source entities produce bad code (have you looked around on Freshmeat lately?) and some commercial software houses produce good code.
How can we tell which is better? Cutting the crap and testing the products in question. If having the source available is important to you, obviously the open source software will win out every time. If only package functionality matters to you, waving a banner won't determine the best choice.
As far as "gold" code, you're wrong. "Gold" code simply means that it's going to be released. It designates that a particular snapshot of the codebase is being burned onto CDs and put into a shrinkwrapped package. "Gold" refers to a baseline, not the quality of the baseline. If you replaced "gold" with "good", I'd agree with you.
- I don't need to go outside, my CRT tan'll do me just fine.