January 29, 2008 at 9:48 am
I'm with Steve. Regardless of the reasons, we should not come to expect people be late in delivering software. Frequent tardiness is merely a sign that Microsoft management isn't capable of properly scoping a project and managing it. Personally, I'm looking forward to features in 2008 that will help me with large scale Business Intelligence projects, but I find myself waiting, and waiting...
I would be less upset if 2008 wasn't the hot topic at PASS. Most Microsoft presentations dealt strictly with features and samples based on the 2008 platform. To get real-world 2005-based content, I had to check to make sure a non-Microsoft employee was delivering the presentation.
January 29, 2008 at 9:49 am
This is going to be controversial but I actually agree with abitrary deadlines, otherwise what date do you work to? Work expands to fill the time you have in which to do it, and a company can't keep throwing money at a development until the design team consider it "perfect", which would be in 2015 if you're lucky. I even suspect that the Marketing department are just as qualified to set a date as the development team - in the same way that a four year old girl picking stocks can outperform an expert stock broker.
What a deadline does is focus the team on the deliverables, whether you hit it or not is kind of irrelevant, if it isn't there you can pretty much guarantee your developers will completely lose focus. They need the urgency of "late" even if the time constraint is largely artificial.
The general theme that seems to be coming out of this is "that's just the way it is", that's certainly how I see it.
January 29, 2008 at 10:26 am
... I actually agree with arbitrary deadlines, otherwise what date do you work to? ...
I don't know about "arbitrary" (companies should develop timeline methodologies which get decent results), but I do agree that companies should have deadlines to work towards. Having a serious, management-watched internal deadline, as opposed to one announced to the public, gives everyone an important goal to work towards.
My problem is when an internal deadline is announced to the world prematurely. What is a legitimate reason to announce a release date to the public before the product is almost done and fully tested? When a company makes an announcement to the public, they are giving their word. If they want their company to have integrity, their word should mean something. That doesn't mean that occasional problems with meeting public deadlines won't occur, but those slip ups should be rare and viewed as most embarrassing.
By the time a company announces a release date, the product should be finished and just going through final testing to fix those last, hard to spot bugs and minor user interface tweaks you didn't know you needed.
January 29, 2008 at 10:53 am
...maybe Microsoft should drink its own KoolAid. After all - Team Foundation Server seems to heavily promote using an Agile development, which among other things, would imply that pretty much at all times, whatever build you're putting out is fully functional. Under that model - just pick a date for releasing the product, and whatever happens to be IN the product is what makes up the product.
..................
Oh - "Do as I say - not as I do", you say?
hmmm........
----------------------------------------------------------------------------------
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?
January 29, 2008 at 10:55 am
Richard Gardner (1/29/2008)
This is going to be controversial but I actually agree with abitrary deadlines, otherwise what date do you work to? Work expands to fill the time you have in which to do it, and a company can't keep throwing money at a development until the design team consider it "perfect", which would be in 2015 if you're lucky. I even suspect that the Marketing department are just as qualified to set a date as the development team - in the same way that a four year old girl picking stocks can outperform an expert stock broker.What a deadline does is focus the team on the deliverables, whether you hit it or not is kind of irrelevant, if it isn't there you can pretty much guarantee your developers will completely lose focus. They need the urgency of "late" even if the time constraint is largely artificial.
The general theme that seems to be coming out of this is "that's just the way it is", that's certainly how I see it.
I truly hope you are joking here. While I agree that if you left the developers determine the delivery date once they decide the product is ready you will end up waiting forever, that is not justification for allowing anyone to arbitrarily set a delivery date. The Marketing and the Sales people as well as a 4 year old are not qualified to do this because they (normally) are not developers and therefore have no knowledge in this field. If you were going in to your local hospital for a heart transplant would you be OK with the administrators arbitrarily setting the date for your surgery? Of course not. Unless those in these departments have working experience/knowledge in programming they have no way of accurately determining a delivery date. The person who set the dates are the project/team leaders and they do this in advance which gives them a date/time to work to and it also gives the sales & marketing people a date to sale/market for.
I have a daughter that will be 4 in less than a year, are you OK with here determining when your next work project is due?
Kindest Regards,
Just say No to Facebook!January 29, 2008 at 11:57 am
jay holovacs (1/29/2008)
Why rush?SQL 2005 is working fine. It will be working fine 6 months from now. I'd much rather the new version be right than soon.
Not everyone is on 2005, and regular support for 2000 is scheduled to end in April. I'd rather upgrade to 2008 when ready, but our business has reservations about staying on an unsupported version for too long.
January 29, 2008 at 12:39 pm
First I want to be clear that I'm not blaming the developers for this. I have contacted some of them, see them in the MVP newsgroups and I think they're working hard and each of them driving to get his section of the product finished.
My complaint is with the arbitrary nature of the marketing department. I don't care if the product releases this year or next, except that support for SQL 2000 ends. And it appears they're not patching it before then. So in that case I want 2008 done. Or support extended.
It's more embarrassing as a SQL professional to have a "launch" event when there's no product. It's scary to have a deadline imposed on the developers if things aren't working. I'd rather have them cut features.
And it's sad that there don't appear to be plans to even provide any more service packs for 2005.
Let the developers build the best product they can and if they can't get everything done, release it with less features. If you're going to release every 2 years, then we can wait for the features.
January 29, 2008 at 2:15 pm
Steve Jones - Editor (1/29/2008)
............Let the developers build the best product they can and if they can't get everything done, release it with less features. If you're going to release every 2 years, then we can wait for the features.
Amen. THere's no need for a revolution in the product every 2 years to keep it it's market share or user base.
Kindest Regards,
Just say No to Facebook!January 29, 2008 at 2:55 pm
I don't see the April 2008 end of mainstream support mentioned any more on the Lifecycles page:
http://support.microsoft.com/lifecycle/?LN=en-us&x=8&y=10&p1=2852
Holly
January 29, 2008 at 3:09 pm
Holly -
Maybe there's something strange going on, but this is what I get when I go to the link you mentioned:
Products Released General Availability Date Mainstream Support Retired Extended Support Retired Service Pack Retired Notes
SQL Server 2000 64-bit Edition 11/30/2000 4/8/2008 4/9/2013
SQL Server 2000 Developer Edition 11/30/2000 4/8/2008 4/9/2013
SQL Server 2000 Enterprise Edition 11/30/2000 4/8/2008 4/9/2013
SQL Server 2000 Standard Edition 11/30/2000 4/8/2008 4/9/2013
SQL Server 2000 Windows CE Edition 2.0 12/16/2002 1/8/2008 1/8/2013
SQL Server 2000 Workgroup Edition 6/1/2005 4/8/2008 4/9/2013
Looks to me that they still haven't moved the date and it's still due to drop on 4/8/2008?
----------------------------------------------------------------------------------
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?
January 29, 2008 at 3:38 pm
2008 has two features that I'm interested in: the breakdown of date/time data types, and the spatial data fields. The former is useful in many applications, the latter benefits my ESRI users, but only once ESRI gets a release out that exploits it.
Our entire installation is SQL 2000. We intend to start upgrading them to 2005 once we get a beefier box. 2000 does quite well for us right now, and since everything is running well, hopefully the end of support (assuming it is not extended) shouldn't hurt us.
Microsoft is definitely too marketing driven. Their ideal for their revenue model is (or at least was) to release a desktop every 3 years or so, an Office release half way between, and a server every 3-5 years. Their marketing people follow this, but the programmers can't keep up. But this is definitely silly with what they're doing for SQL 2008, announcing the product months (if not a year) before it will actually appear. Most of us know better and apply a hefty amount of salt to their statements.
It would be enough to hear "MS developers are working on the next version of SQL Server and expect the initial CTP release soon." We don't need formal rollout announcements months before it goes RTM.
-----
[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]
January 29, 2008 at 11:46 pm
I have spent the last two months converting a huge in-house ecommerce application written in 1998 from mssql 6.5 to 2000. The migration has took off during the holydays when traffic was slow in the office and it was very painfull. Why did we upgraded an application that was working just fine? Because we want to move to MSSQL2005 eventually and keep up with licencing.
I am working in a small business of about 30 employees, but controlling most of its market share's of one province in Canada. You can bet on which it is by my bad english.
Anyway, as a small business developper, I think it is difficult to get along with upgrades every 2-3 years. We are always working with old technologies but trying to keep up to date.
MS is not th worst company in that field. My own employer is a VAR for Autodesk that releases a new version of their product in all verticals every year!
I can tell you that sometimes I wonder why a few features deserves a new major release.
January 30, 2008 at 6:28 am
Yes, at that link it does give the 4/8/08 date for the various editions, but for SQL 2000 SP4 it does not, it says until next service pack or the end of the product's lifecycle, and under support lifecycle they say "Mainstream Support for Business and Developer products will be provided for 5 years or for 2 years after the successor product (N+1) is released, whichever is longer." So it seems like that would mean 5 years from 5/26/05, the general availability date given for SQL 2000 SP4. ?? Everyone tells me, no, it is moving to extended support 4/8/08, so I must be missing something.
Holly
January 30, 2008 at 1:09 pm
If it ain't broke, don't fix it!
If you are running SQL2000 and your boxes aren't stretched then where is the pressure to upgrade?
There are some features of SQL2005 that really boost performance. Partitioning for one. The average execution plan seems to have half the cost.
The feature that I am interested in for SQL2008 is the resource governor and the spatial data types. I have to say that intellisense in the RTM version is a way behind Red-Gate SQLPrompt. Actually, it appears to be behind the pre-Red-Gate release which was PromptSQL!
Is anyone using the service broker in a production environment?
January 30, 2008 at 1:41 pm
Microsoft is pretty quick to drop support on prior versions.
Not every vendor does that:
http://h71000.www7.hp.com/openvms/openvms_supportchart.html
OpenVMS VAX version 5.5-2 released in Nov 1991
Ending date for Prior Version Support-Sustaining Engineering:
At least through 2012, with 24 month notice thereafter
Viewing 15 posts - 16 through 30 (of 33 total)
You must be logged in to reply to this topic. Login to reply