Slashdot Mirror
New SQL Server Release Slips to 2005
Strudelkugel writes "CRN reports SQL Server 'Yukon' will slip to 2005, complicating plans for ISVs and creating opportunities for OSS and other competitors."
← Back to Stories (view on slashdot.org)
← Back to Stories (view on slashdot.org)
Has any one contemplated the concept that Microsoft might actually be taking the time to make better products? I realise its taboo on slashdot to show any support to Microsoft, but the fact is that they are not stupid! Do you honestly believe they would just decide, hey, lets let linux + competitors get a foothold in our markets whilst we jack about! WRONG!
One thing anyone in the IT business should learn is to never ever under estimate microsoft.
As a DBA who deals with MS SQL 2k (and 7 and 6.5) on a day to day basis (hour to hour basis?) I'm actually kind of saddened by this. I was really looking forward to playing with the TSQL/.Net paradigm shift as far as accessing data.
:)
7.0 was a huge jump from 6.5 and 2k from 7.0 was almost as significant of a jump. I will call a spade a spade and say that the evolution of the MS SQL server has really impressed me and I was looking for good things from this next version as well. I know this is the wrong place to say such things, but I've had lots of problems with other MS problems, but this one since 7.0 has been quite good. Don't even get me started on some of their other products though.
I'll just go hide in my DBA hole until 2005 I guess.
--Mike--
If I had any doubts that MS is helping OSS and slowly erasing itself, it is now clearing :-)
Jokes aside, this will seriously affect businesses that have paid for their upgrade licenses, as the licenses will expire before the sql server is released. This will make decision makers view Open Source in a new light. Atleast, in Open Source you don't pay for future vaporware in the present.
"In questions of science the authority of a thousand is not worth the humble reasoning of a single individual."
I saw a couple of Alpha builds of Yukon and the Planning papers (blue badge), but I didn't see much, but I bet I know what's taking so long:
.NET types and SQL types automatically, and allow you to run C# SQLDataAdapter-type code natively within Stored procedures. Plus with the trend starting in SQL 2000, it'll be XML, XML, XML. I know XML will be a native type and some of the "indexed xml" (red/blue fast-search vs. DOM-search) that they started in the aborted Hailstorm project will be in there.
.NET; Yukon replaces the SQL you knew with new stuff. They'll eventually get it right and it will rock, but don't expect to use all this until 2007 (it'll be out before then, but you won't finish your first REAL project till then).
Yukon will allow structs as column types, and will do mapping between
Longhorn replaces Win32 with
There, I said it. 2007.
Many business customers have recently been coerced into signing ongoing contracts where they receive any upgrades in a particular year in exchange for a yearly fee.
These companies are going to be extremely p155ed off when they realise that all they are going to get for their money is (maybe) XP Reloaded (think ME).
Companies cannot afford to throw money down the microsoft toilet for much longer... especially when all they get is extra bugs that they didnt need in the first place, coupled with a healthy dose of lock-in and increased support costs.
i think you'll find PostgreSQL is also pretty good value for money!
I think they are delaying not due to stealing OSS software ideas, but honestly trying to make their software better. For the first time in many years MS has real competition. They can't release another insecure trashbag OS or database server. If they release before it is actually ready, then they will get tons of bad press and their lunch ate by OSS software.
Lets just hope OSS developers don't sit on their laurels during these delays. If they do they will be playing major catch up come 2005/2006. This is the time for OSS to take the lead. The boys at Redmond may be evil, but they are no fools.
This is one of Microsoft's most important products. Finacially, there is a huge amount of "positive perception" riding on SQL server.
Businesses may run on one of their OSes, but businesses run IN SQL Server. This product can make or (more critically) brake businesses. If rumors of major problems with SQL server screwing up business were to get out, corporate perception of them would tank.
They have no real choice with this product but to try and make sure it is ready (and take more time if needed) rather than push it to market.
-Pete
Soccer Goal Plans
It is really funny the level of fervor behind Mysql. So funny it makes you wonder if the zealots have ever used anything other to any real extent.
The company I work for software's backend can go Mysql, Postgres, Mssql, Db2, or Oracle.
For massivce connections, queries, reporting, reliability it is in this order.
1. Mssql, DB2, Oracle, all pretty much equal.
2, Postegres, tricky but holds its own.
3. Mysql, will work in the low end, forget reporting, forget huge db hits.
I like Mysql. But Mssql 7.0 hands its ass to it.
What happens is some company will be our product. Hand it over to some 25 year old self proclaimed web genius to install. Conversation is as follows.
1. "Can I have the Source?" No, it is closed, long discussion about how we suck cause our product isn't open source.
2. "Ewwww, Java, it sucks, you should rewrite in PHP" I explain it has been continually developed since 96, no way to stop the engine and write in PHP.
4."I decided to save the company some money and install Mysql" We say ok, explain issues, put them in an email and fax(CYA principle). I then advise to run Postegre, that it is more robust, and is FREE as well.
No one lists. Junior installs on Mysql, everything runs fine, site gets huge amount of traffic, database gets quirky. Management starts running huge queries on database reporting tool. Database is very slow to respond, then in a few weeks keels over.
We get called. Tech is yelling, my guys are smirking(but still polite on phone) Management, myself, and tech gets on conference. Tech starts berating me. Management starts berating me. I pull out magic email and fax with all my system recquirements, suggestions for optimal use. Hey, guess what I was write. Wait a minute, shouldn't I know best since I work for the company that writes and support the product?
Three times a week this happens with Mysql. We have 14000 customers and I swear 50 percent have some guy that thinks he knows best.... knows our product better, knows computers better...
This is a great example of where our community needs to clean up its act. And I thought I would never say that.
Mysql is good for what it is, but there are many things it is not. Learn this.
Puto
The Revolution Will Not Be Televised
How about ANSI '92 compliance for MySQL... that would be a good start!
No, a good start would be to flush MySQL down the toilet where it belongs and use a real database engine such as PostgreSQL or Firebird.
Seriously! Why wait for MySQL to add all those missing features when such superior alternatives already exist, and, furthermore, MySQL has a more restrictive license?
Give MS a frickin' break....MS said there is going to be something like 40 *million* lines of code...
Just out of curiosity, I counted the lines of code (both c & assembler, all processors) of the 2.6.4 kernel. It is less than 5.5 million.
40 million lines of code. There's all the reason I ever need to not use it.
With 40 million lines of code, you never fix bugs, the best you can hope for is to relocate them to a really obscure place.
Suppose you were an idiot. And suppose you were a member of congress. But then I repeat myself. -- Mark Twain
As long as you can accept the limitations of MySQL, it's perfectly usable. MySQL is faster and lighter weight than PostgreSQL in my experience. I haven't tried Firebird yet.
Honestly, I wouldn't want to run a site like Slashdot on MySQL, but for smaller projects it seems useful.
RDBMSes don't implement Codd's 12 rules anyway, so maybe none of them are "real". Personally I think it's good to have a range of database options. At the high end, Oracle and DB2 have loads of features, and are presumably "real" by your definition, but they are also incredibly complex to administrate, which is why most companies have dedicated DBAs for them.
Oh Joy
Just what you need a new microsoft database that makes refactoring and porting your DB to another platform near impossible.
Larry Elison is probably chuckling like a demented monkey over this. I can see his sales people going at this. Microsoft Software assurance = Pay them to take their time to devise ways to achieve complete customer lock in. Or, the ever popular why run your business using techniques with 50 years of validation behind them when you can do things microsofts way.
I can allready see the security problems popping up. Run C# code directly, the same code being ever more integrated into yukon. Well seems we will be able to expect worms that make slammer look like a joke. Heck you could have them replicate throughout the entire system and hold entire enterprises data hostage.
The sad thing is that the large group of IT director/ Sysadmin lemmings will go along with no one ever got fired for choosing microsoft. After all, look at how they have embraced the ever popular and ever more dangerous office/exchange combo.
SQL Server 2005? is going to be great. However, if I had to choose the *best* database I would go with Oracle without a doubt. Every tool other database manufacturers are trying to mirror generally come from Oracle. Plus they support Linux which makes buying very expensive hardware a problem of the past. Hell you can get a license for standard for $799.
Unfortunately my job runs SQL Server 2000. Having cut my teeth on PL/SQL, Transact is a nightmare because it is so limiting.
I'm actually looking forward to Yukon because the marketing ad sheet shows some really cool features. The only question is will they deliver and when will it be?
Much as I love a good MS Bashing, I'll tell you what I find really lacking (personally) for PostgreSQL and other OSS RDBMSs - a good GUI management tool.
Something that helps you craft medium-complicated joins quickly with a few clicks and drags.
For example, see this screenshot from Visual Interdev working on MSSQL2k, creating a SQL Query for a stored proc. Sure, it's almost trivial to hand-write the SQL code. But it was even easier to just select a few tables, click on the fields I want, right-click on the joins (created automatically from the database structure) to change their type, and be done.
I use PGSQL for all my personal projects now, but I sorely miss the speed that a GUI editor like this allowed me.
The real problem is not so much that the Yukon date has slipped, it's that Whidbey (The next version of Visual Studio.NET and the .NET framework) is slipping with it. For who knows what reason, Microsoft decided that these products must be released together. While Yukon promises some very nice features, most people would much rather have Whidbey released now and live with SQL 2000 for awhile longer.
To top it off, MS is not even going to be releasing any service packs for Visual Studio in the meantime. There are some rather serious issues with the current version of Visual Studio that can only be fixed by calling MS for specific hotfixes. Needless to say, much of the MS developer community is up in arms.
MySQL...is ten times easier to manage and work with then SQL Server 2000.
I'm sorry, but what the hell are you talking about? I've used both these servers extensively (as well as Sybase ASA, PostgreSQL and Oracle), and as much as I respect MySQL, it's certainly no easier to use than SQL Server. It's at best about the same, with SQL Server being much easier to pick up from 0 knowledge due to a surprisingly good set of help docs. Enterprise Manager and Query Analyzer are really good tools, as well...in fact, until we discovered mssqlXpress, Query Analyzer was bar none my favorite IDE for making new statements. (sqlXpress adds sourcesafe integration, versioning, and historical reporting to a clone of Q.A. with autocomplete and automatic proc generation, it is a pretty clutch tool)
MySQL is very good, but ten times better? Not really. In fact, if I had to beg for any SQL Server regardless of price, I'd take SQL Server because it's the easiest to develop for and easiest to port FROM. This gives you an app that will run on almost any other server with a little effort. I rewrote a massive app to run on Sybase in three weeks and Postgres in a month (most of which was testing the DB core of our app).
Hey freaks: now you're ju
As technically inferior MySQL is to Postgres, MySQL has a few major things going for it that ensure it's niche.
1. Easy to install on Windows. The average coder at a Windows-only farm can easily run the executable and have the latest version running on their developer box. Not all companies allow you to have multiple boxes, and many force you (via draconion security measures) to only run windows with certain software installed. Postgres NEEDS a user-friendly Win32 installer, perhaps with a similar info-item like MySQL has. This is a MUST for companies to start to take notice. Then, a PHB can even play with it and like it.
2. Marketing. While open-source, MySQL has a nice marketing engine behind it. A beautiful webpage, online and PRINT adds, and magazine and newspaper articles CONSTANTLY writing about the "little database that could" every few week / months. Postgres needs to start getting the word out, and hype it a little. Just because a product is superior, doesn't mean it will thrive. There are tons of examples out there: Beta vs VHS, Windows vs OS X, etc. For a database to be used, it must be allowed and "signed off" by a manager of some sort. Most will take reputation + support + "ooh, nice webpage" over a product that might be better, but they know nothing about it.
3. More management tools. MySQL has a couple out there that look and run great; very professional looking. This earns respect from PHB's, as they are easily misled by such niceties.
Don't get me wrong. MySQL is nice, but doesn't have what I need most (Views, triggers, etc). Postgres may not be perfect, but I think it is superior. We just need to get the word out to those "not in the know".