The Social Contract

  • Comments posted to this topic are about the item The Social Contract

  • Heh... If I remember correctly, it's called "Agile" programming... and it sucks. I no longer wait for sp1 to upgrade... now I wait for sp2.

    --Jeff Moden


    RBAR is pronounced "ree-bar" and is a "Modenism" for Row-By-Agonizing-Row.
    First step towards the paradigm shift of writing Set Based code:
    ________Stop thinking about what you want to do to a ROW... think, instead, of what you want to do to a COLUMN.
    "Change is inevitable... change for the better is not".

    Helpful Links:
    How to post code problems
    How to Post Performance Problems
    Create a Tally Function (fnTally)
    Intro to Tally Tables and Functions

  • Hmmm. So far SQL 2005 isn't that far behind SQL 2000 in terms of service packs issued per year.

    It'd be nice to see the cumulative update pack 6 more thoroughly tested and turned into SP3. I'm just getting my feet wet with analysis services but it seems that it is disproportionately represented in cumulative update packs in terms of number of fixes... Thus I've applied CU6 just to be sure that I don't run in to some problems. But then of course I have the issue of having to ensure that my development environment functions the same way as various customer production environments... SP3 would make this much more straightforward 🙂

  • I'm still not sure that "should" and "shouldn't" play much part here. Microsoft sells tools to do jobs. Companies and/or individuals buy the tools they need. The closer the match between the tool's capabilities and the need being filled at a given price (initial purchase + "running costs"), the more likely the consumer will come back to Microsoft and buy more tools, either as replacements, upgrades or for new needs.

    If Microsoft degrade what they're offering, they'll lose customers. Linux wouldn't have gained the foothold it has if Microsoft had trodden a little lighter in the OS market (IMHO, that is). SQL Server has some very good competitors, so many current customers are far less locked into MS products than MS would like. So eventually, Microsoft can do what they like with SP3 for SQL Server 2005, but we customers can also do what we like with our money. And from my point of view, if the cost of maintaining SQL Server in whatever flavour ends up too expensive, I'll move platforms.

    It's all only brinkmanship in another wrapper.

    Semper in excretia, sumus solum profundum variat

  • This is so true about the won't upgrade until SP1 culture, I gave MS benefit of the doubt with Vista and got bitten. SP1 does now seems like the product they should of released in the first place.

    When SQL 2005 came out we only installed it on development machines and wouldn't release anything into production until SP1 was on the streets for a good month (just look at the SP2, SP2a problem).

  • Why would anyone think it could possibly be OK not to fix manufacturers defects in a product?

  • Bob Matthews (3/11/2008)


    Why would anyone think it could possibly be OK not to fix manufacturers defects in a product?

    Because OK is a variable term. Generally, manufacturers don't do things for your benefit; they do things that provide them with a route into your wallet i.e. their benefit. I don't like it that way, but understanding it saves me wasting effort kicking against something I can't change and allows me to channel that effort towards more vulnerable targets.

    Microsoft won't react to arguments of "please provide this service pack because it's your moral obligation". They will, however, react to arguments of "please provide this service pack or you'll lose my future business". There's no right or wrong, just a power struggle between a powerful manufacturer and a large (and wealthy) customer base.

    Semper in excretia, sumus solum profundum variat

  • It's not quite as simple as we'll buy something else. Many companies have expertise in areas and it's a big decision to switch technologies as people may not want to, many may not achieve the skill levels with the next product, etc. So in some sense, Microsoft can count on this in the medium term, a few years, but in the long term, I'd agree. If quality suffers in SQL 2008, 2011, Windows 7, etc., I can see quite a few people looking for alternatives. Especially if price pressure for Oracle/DB2 brings them more in line with SQL Server.

    Course, not sure those products are any less buggy. At least I see IBM releasing service packs on a more regular basis.

    We are in the third year of SQL Server 2005, which should bring a new SP. However it isn't, so we'll be in the 4th year before we see it and given a quick release cycle, it might be the last one we get.

    We do need to vote with our wallets and our voices. So do we boycot 2008 until 2005 is more stable? Are you willing to add some votes to Connect to show them that we want this? Maybe pass the word at all Code Camps that we should be calling for SP3 and a regular patch cycle?

    Do you think we should get regular SPs? You might not have had trouble with the CUs, but many people have. Or won't install them because the warnings indicate it hasn't been thoroughly regression tested.

  • I guess i look at it a little differently. They are on a 3 year release cycle now. That tells me that really the new versions are not going to add massive changes but really implement what they just didn't have time for in the previous one. SQL Server 2005 we actually recieved a partially functional app as the initial release. SP1 gave us Mirroring support among other fixes. SP2 fixed other core functionality. I would rather they put their energy into a stable 2008 and make it so we don't need Service Packs as regularly than to work on what is nearly legacy code. Once they have a stable release, I do think it wise to go and do one final wrapped up update which is tested. Most likely just all of the customer updates they have created for customers wrapped into a single cumulative update. If you are suffering from an issue apply the update or get them to write one before you go into extended support. If you just want an update for the sake of an update then what purpose does it serve. There are tons of SQL servers that are on a base install or 1 sp up and they function at 100% for their use.

    Are we asking for a wrapped up SP for the sake of having one or because we are seeing issues across the board.

  • Software Assurance takes on a whole new meaning if you are expected to upgrade as soon as the next version is released.

    Seems Visual Studio also suffers from a similar fate - expectation of another SP that did not come as the next version was going to be released.

    SP2 as released was broken for us (partitioned cubes :crying:), I had to wait and 'find' the right CU that would work. The CU's are not promoted like SP's. Many thanks for the ones trying to keep track of all the different release levels.

    They should do an SP3 to at least try to roll up 2005. I don't think most companies want to feel like they're part of an extended development environment.

    Greg E

  • The place where the social contract seems to be broken is that compared to 2000, the licensing prices quadrupled, and for a product that has "half the shelf life". Like it or not - 5,000 was the per proc (retail) price for EE in 2000, which if my math is correct (I need both shoes off for this) - would last you for 6 years, or 5 and a half if you upgraded to 2005 on day 1. Now - we're expected to shell out 20K for that same license, for something lasting half that time.

    And quite honestly - for a product that was not all that more stable.

    I know all of the arguments about "so much cheaper than Oracle", etc... but still - for that much of an increase (which in my mind is working out to an eight-fold increase per year), damn it - there should be a substantial improvement in stability AND performance. Out of the gate.

    Taking a year to put out SP1 on a product that's around for 6 years means you get to leverage 85% of the product's effective life. 1 year out of a 3-year cycle means you're squandering 1/3 of the shelf life. I'm sorry - but OUCH......

    ----------------------------------------------------------------------------------
    Your lack of planning does not constitute an emergency on my part...unless you're my manager...or a director and above...or a really loud-spoken end-user..All right - what was my emergency again?

  • I don't agree with all the views presented here, but that certainly doesnt make them less valid! But if you back way way up, here are our choices:

    - Do nothing. Apply whatever service packs they push, and hunt for hot fixes if you run into the problem. Entirely pragmatic

    - Do something. Vote for SQL 2005 SP3 with a view towards making MS commit to some policy about service packs through product end of life

    MS isn't perfect, SQL Server isn't perfect. Doesn't mean it's a waste of time to speak up about things we'd like changed, features we'd like to see, etc. So far maybe 300 people have cared enough about the issue to vote for SP3. I'd be surprised if MS changes course for 300 people given the size of the SQL user base. At 3000 I bet it gets interesting. At 30,000 I can MS jumping through some hoops to get it done.

  • As far as feedback goes, especially with Connect being the main source of bugs/issues with all MS products, here are a few items.

    Items (all status, open or closed) with > 100 votes: 15

    Items (all status, open or closed) with > 100 votes: 4

    Items (all status, open or closed) with > 100 votes: 3

    Items (all status, open or closed) with > 300 votes: 1

    SP3, 563 votes as of now

    This is in SQL Server. In Windows 2008 Server, there are no items with 100 votes or more. There's no Vista area or other popular products that I've expect to see. No 100 vote items in MOM/Op Center either.

    So we're showing them this is important, at least to lots of customers.

    So Vote!!!! (https://connect.microsoft.com/SQLServer/feedback/ViewFeedback.aspx?FeedbackID=326575)

  • I'm not sure I agree that the value or enhancements with 2005 aren't significant for the price increase. Don't forget inflation as well, so over 5 years, I'd expect some price increase.

    SS2K5 added a lot of features and functionality. There are tons of things that are added (CLR, Service Broker, substantial SSRS changes, new T-SQL like TRY..CATCH, etc.) that can add value. However you have to build applications to use that stuff. My feeling is that most people run COTS, which means that none of these are useful and why move to 2005?

    Scale is one reason, ss2k5 runs faster, has better tuning, etc., but you need to be sure this is a good value for you. 64-bit support is better, but still flaky in places. This is one place (if you have large data sets/sizes, and scale issues) where 2008 might make sense.

    More stable. That's a good point because for most people (no clusters, running Standard), SS2K is very stable. As in 99% stable, so is it worth lots of $$$$ to get to 99.9? Getting above that is hard with Windows patches and human errors. And costs a lot.

    Is SS2K8 a patched SS2K5? I'd argue yes, in that v6.5 was a patched v6.0, 2000 was a patched v7, so it appears that from past history, the big rewrites were good, but they built a new 1.0 product. While SS2K5 is a great product, in some ways it's a 1.0 platform and many people are looking to take advantage of those new features. And figuring out what works and doesn't. so in some sense, while it has lots of new features, SS2K8 is a patched SS2K5 with the core changes in 2005. The new features, like Resource Governer, Policy Management, etc., are 1.0 features.

    Maybe that's the new idea. Each version will have 1.0 features, which you want to be careful of, and then 2.0 versions of the 1.0 features from the previous version.

    Perhaps I need a "Use, but be careful" and a "use as you can" list of features for each version 😉

Viewing 14 posts - 1 through 13 (of 13 total)

You must be logged in to reply to this topic. Login to reply