May 12, 2008 at 7:26 am
Upgrading is expensive and possibly problematic. There is no smoother transition than no transition. We are using SS2005 where I work, and there is no money for upgrades.
Although there are nice-to-have features in SS2008, for our business, there are no necessary-to-have features. At least, I haven't convinced my boss of this (yet :P).
As a technical-minded individual, I always want to learn the new thing and play with new technology. On the other hand, as an individual working within a business structure, there is a certain bean-counter that would need a serious amount of proof-of-need before parting with any amount of cash.
I don't get to choose whether to boycott or not. If I could choose, I would choose not to boycott. There is no real way to keep new technology down or to stop capitalism in the USA. The only way to curb things is by law. If there is some technical lawyer out there that wants to force large companies to be responsible for their product, go for it. How long do you think that large corporations would allow fiscal irresponsibility if it were not for Sarbanes-Oxley?
Mia
I have come to the conclusion that the top man has one principle responsibility: to provide an atmosphere in which creative mavericks can do useful work.
-- David M. Ogilvy
May 12, 2008 at 7:39 am
Good comments and I'm not really calling for a boycott of 2008. I wasn't sure how to tackle this and was looking to see how others feel. I like the changes to the support policy and in many ways I'd still suggest upgrading to 2008 instead of 2005 since they are very similar in how they work and 2008 is probably more stable if you don't use the 2008 features.
The hot fixes are annoying, but they are all roll ups. Each of them contains the other fixes before it and. There are cumulative updates every other month and they can be applied on top of SP2 and contain all updates before that one. So CU7 has CU6, CU5, etc. inside it. Right now if you are running SS2K25, you apply SP2 and CU7 and you are up to date.
Microsoft probably won't hire me, and I don't really want to leave Denver, so I thought I'd put this out there since I it could have merit. I'm not sure the business model for Microsoft would change. I do think that all the resources on ss2k8 is a bad idea. To me there should be a good portion (10-20%) of developers working on 2011, either designing the future or trying to find ways to make current things more stable.
May 12, 2008 at 7:46 am
We don't touch new technology until it's proven and settled (a service pack or two) and passes the acceptance testing and also has good reason to move to. We are only just moving to SQL2005 so are unlikely to go to 2008 until 2005 is out of support or a glaring security hole appears. So yes effectively we are boycotting 2008 as we just aren't interested yet.
Ditto. Plus, I have to wait until my vendor certifies their product on 2008. They just certified their product on 2005 a few months ago.
May 12, 2008 at 7:47 am
I do have a friend who does contract work on SQL and will be moving to 2008 as soon as it goes RTM. There are a number of new features that he indicates will solve many of the problems they are experiencing now.
May 12, 2008 at 8:05 am
There's no way that a shop such as ours with several dozen different dbs and apps and only 3 DBAs can perform upgrades every 3 years. We still have a couple of SQL2005 conversions left to do.
As to the boycott - we may start playing with SQL2008 in a small test environment later this year, but we probably won't even consider a conversion until the second half of 2009. And that largely depends upon finances, too.
I think a 5 year cycle for major releases is more realistic.
May 12, 2008 at 8:38 am
If you're looking at a 5 year cycle, then would you skip versions? Go 2005 -> 2011?
May 12, 2008 at 8:45 am
I don't think we'd skip a version. I just don't see how we can upgrade every 3 years with our current level of staffing. We would always wait several months after a release is issued before giving it any serious consideration. We can't afford to be on the bleeding-edge. I guess we'd just lag behind by a year or two.
May 12, 2008 at 8:51 am
Not to mention cost. Unless you have Software Assurance you have to repurchase SQL Server every three years. Unless you can really take advantage of new features I don't see how many companies can cost justify it. We cannot go with Software assurance mainly because 90% of our software is purchased and we have no way of knowing when the vendors will certify the new releases. More than half still have not certified SQL 2005. Add to the fact that our servers and SQL licenses are depreciated on a 5 year cycle.
To me Microsoft is turning SQL Server into like what we see in Office upgrades. The benefit of upgrading is starting to be smaller and smaller and harder to justify upgrading. Not to mention more and more 24X7 critical systems are running on SQL Server and we just cannot just upgrade because it is new and risk our jobs.
May 12, 2008 at 8:53 am
I 2nd P. Jones -
"We don't touch new technology until it's proven and settled (a service pack or two) ... So yes effectively we are boycotting 2008 as we just aren't interested yet. "
May 12, 2008 at 8:54 am
All of my servers: 14 instances on 8 boxes, 20+ databases; are 2000. I'd like to get them to 2005 just in case we have a support issue, but it's not critical as all of my boxes run pretty smoothly except for one that's on horrifically underpowered hardware. I'm interested in 2008 for two features: spatial data types and (FINALLY!) the separation of date/time data types into date and time.
The former gives benefit to my GIS systems once ESRI gets their systems updated (and they've demonstrated on 2008, so it can't be too far away, can it?), the latter gives benefit to new application development as it's rare that we need the time component.
But will I boycott? Not directly. I won't upgrade until SP1 is out for SS2008 and I get word from other ArcInfo users that their SS2008 version is working well. I have no need or interest to go to 2008 at this time, and since I'm the only DBA that we have, I have ZERO interest in being on the bleeding edge!
-----
[font="Arial"]Knowledge is of two kinds. We know a subject ourselves or we know where we can find information upon it. --Samuel Johnson[/font]
May 12, 2008 at 8:57 am
Phil Factor (5/12/2008)
If Microsoft claim that they have lost interest in maintaining and developing SQL Server 2005, maybe we ought to suggest to them, with a straight face, that they can turn it into a community project like PostgreSQL. When you see their reaction to that idea, you'll soon see that their lack of interest in 2005 is just a ploy to force the industry to upgrade.
Gee, Phil, I can see the headlines the next day: "Bloodbath at Microsoft: 50% of SQL Server developers and C-level directors die from cerebral explosions!"
Do you really want that many people to go all splody? 😀
(on the other hand, it might make for a heck of a You Tube video!)
-----
[font="Arial"]Knowledge is of two kinds. We know a subject ourselves or we know where we can find information upon it. --Samuel Johnson[/font]
May 12, 2008 at 9:18 am
Hello,
I agree partly with you.
I think that's easy to upgrade if the new developements are done in a careful way.
1) the problem with modifications done by Microsoft about the specifications of Sql Server is : we have to modify the instances, the databases and the programs which are using these databases
2) never to use deprecated features in the current version of Sql Server ( it seems that the support of any product will stop with the disposal of the second version )
3) the rule for me ( when it's possible but i think that's only a question of having the heart to use all the measures necessary to do it ) is for new developements to rely on new features of the last current version and in any case to use already deprecated features. For old projects, to seize the opportunity of modifications to clean from the projects everything which is deprecated
4) between 2 versions, we could hope 6 upto 7 years. That would be enough to train teams to the new changes and to modify the programs
5) It's necessary to have a small team specialized in the "chase" of evolutions, which will have to test these evolutions and to foresee all the consequences of these evolutions.
6) What are you thinking about a software company which has succeeded to pass his dos-application to a windows application in 2005 ? Too complicated and useless were the answers of this company ( 70 persons were working in it in 2005 ). Now this company has 12 persons and is going bankrupt.
I know that more important is a company, more difficult is to change the things. I've worked in an insurance company. In 1998, the project "chief" was dismissed because he wanted to create a new table with 6 dates including century : too costly in disk place. At this time, the estimate of change of the date format was known : 15 man years.Stupid.
It's essential to follow the modifications of a database engine and to foresee their consequences.
It's also essential to avoid to rely on old solutions : we must evolve and we must be trained for that and we must prepare the managers to admit that ( and it's not easy : money is money )
Have a nice day
May 12, 2008 at 9:23 am
We like others let the business drive the decision to upgrade. We have not and probably will not boycott any software product. However like Vista it could be years before we move our production environment to it.
More simply put we are a few months away from going live with Office 2007 for all users. The cost and frequency of moving from version to version is high and we do not see the need to move simply because a new release is there. It all has to be justified.
Good discussion!
Not all gray hairs are Dinosaurs!
May 12, 2008 at 9:48 am
From my best estimate, the latest [publicly available] hotfix is build 9.00.3159 (SP2 + Q934459) while the latest [PSS Only] hotfix is up to 9.00.3239 (SP2 + Q949095 aka 'Cumulative HotFix 7'). Should they release a Service Pack 3? Yes.
Geez. If we have to pressure MS to make a cumulative hotfix publicly available because they're not willing to release a service pack, something's wrong with the process.
May 12, 2008 at 9:59 am
We're a very small company running only 1 instance of SS2005. We may not upgrade to 2011 unless we have to. And being as I am the entire IT department, I want as little difficulty as possible!
Viewing 15 posts - 16 through 30 (of 45 total)
You must be logged in to reply to this topic. Login to reply