December 3, 2008 at 8:54 pm
Comments posted to this topic are about the item Just Keep Going and Going and Going
December 4, 2008 at 3:50 am
Hi Steve
Thought provoking words from the editor as usual.
In the UK, many large organizations have adopted an outsourcing strategy. This can range from the simple release of the support & maintenance of legacy applications through to the extreme wholesale release of the entire IT infrastructure and it's staff.
Unfortunately, the extreme case is quite commonplace. I have worked for 3 companies who have chosen to outsource their IT function and have also worked for 2 large outsourcing companies. In all cases, the same points apply:
Outsourcing is primarily driven by bottom-line costs. It is not often that a company will outsource with the belief that the outsourcing company is`better at IT` or will provide an improved service. Outsourcing occurs because the organization believes they can reduce costs and still supply a satisfactory service. There is little consideration for the `practicalities` of development, maintenance and support.
Once outsourced, the outsourcing company must achieve the cost savings declared to the customer. They are allowed to realign resources, reduce headcount and off-shore the work. The transition of work to off-shore resources can take years and all occurs whilst normal `Business As Usual` service is maintained.
An abundance of challenges and difficulties arise here, but my point is this:
Whilst the transition is in progress, do you think software upgrades and patches are high on the list of business priorities?
December 4, 2008 at 5:57 am
If Misrosoft offer hosting to comercial companies, I suspect they will discover some of the realities of the commercial world. Having worked for a hosting provider that also offered management services (OS, DBA, etc), I can say that the customer is king.
Very few customers are willing to pay for an upgrade just because it is available. They need a business incentive - such as reduced hardware, software and administration costs - to outweigh the business risk of making a change. Some customers are very intolerant of risk - my understanding is the IBM are (or were until recently) hosting services for a particular bank that was using software released in the mid 1980s. The customer felt that paying for support for the obsolete software was less risk than upgrading. Other customers are willing to exploit new software as they can see it will help give them a business advantage, but this does not always mean they will move every existing production system to the new stuff.
Even if Microsoft start out today only offering to host SQL Server 2008 and above, I suspect that in 2020 they will still have a number of SQL Server 2008 instances being hosted. Their only other choices are terminating the contract or offering to pay the upgrade costs - neither of which are as financially attractive as taking the money and supporting what by then will be a very low cost platform.
Original author: https://github.com/SQL-FineBuild/Common/wiki/ 1-click install and best practice configuration of SQL Server 2019, 2017 2016, 2014, 2012, 2008 R2, 2008 and 2005.
When I give food to the poor they call me a saint. When I ask why they are poor they call me a communist - Archbishop Hélder Câmara
December 4, 2008 at 6:37 am
I doubt our company would ever consider outsourcing, simply because of security concerns. However, the point about not upgrading because of application breaking needs to be addressed.
Best practice is to be conservative in coding practice. Beyond the development staples of clear elegant code, lots of comments, etc, there is also a future-proofing consideration.
Don't use things outside the core like third-party code addons. They give you short-term convenience at the cost of long-term pain. MS in particular is very very good about backward compatibility (thank God) so there's usually plenty of time if you have to work around deprecated features. There aren't that many and they don't happen that often.
But stay conservative. After all, they call it "bleeding edge" for a reason...
December 4, 2008 at 10:38 am
Good comments, and I wonder how well Microsoft would enforce upgrade requirements. It will be interesting to see how this plays out and I would love to know if Energizer is actually maintaining any SQL 2000 instances or being pushed to 2005/2008.
Staying conservative is good, but if you use something like Dynamics/Siebol/etc., you are somewhat at the mercy of those companies supporting new versions/SPs as well as paying for them if needed.
I'm not sure this is a great move for SQL, or perhaps even Exchange. And if it's just file servers does that make sense? Or maybe it makes sense just to let someone else manage the data center, everything from the Transport layer down, and still keep staff to handle the application level stuff?
I wonder how many companies will think about this.
December 4, 2008 at 11:08 am
Microsoft probably would support a legacy system, but for the right price. A couple of posts to the WSUS forum recently for one of the nastier security vulnerabilities seemed to indicate folks still have agreements in place for NT 4.0 support. That was EOL a while ago. I remember talking with our TAM about that when NT 4.0 was finally going to be dropped unless you had a support contract explicitly for that purpose. The cost to maintain were rather high.
K. Brian Kelley
@kbriankelley
December 4, 2008 at 1:30 pm
K. Brian Kelley (12/4/2008)
Microsoft probably would support a legacy system, but for the right price.
I agree. It makes sense to me that Microsoft could profit handsomely by supporting their legacy systems for customers that were willing to pay for it.
Paul DB
December 4, 2008 at 3:12 pm
Great thoughts Steve. Yet another caveat to potential outsourcing. Having hosted various platforms for web hosting, I have experienced the sudden death of some existing features that was not expected.
I reckon this happens whether it is hosted local or outsourced, it just has more roll-back control when it is local, and more immediate response.
I remember the wild changes between PHP versions when they switched to the Zend engine... what a pain, especially with image handling functions.
Don't ever expect an outsource to do your homework for you. Systems and apps (whether outsourced or not) are still in-house responsibility, as the in-house bottom line depends on it. If not, then the in-house soon lands in the out-house.
February 2, 2009 at 9:56 am
One problem I have with hosting is something we have just gone trough here, and is one of the reasons I just now have time to go trough my emails from SQLServerCentral 😉
We had outsourced the hosting of our CRM software, when the relationship with that vendor went sour, we had a hard time getting a copy of our data (other issues here, such as the marketing department getting the CRM software and not having a very good contract). For me that would be more of a nightmare than upgrades etc.
That is of course not so much an issue if you just outsourced your maintenance, but kept servers in house. But, after being maintained by MS for 2-3 years, who on staff would be able to get to the data?
February 4, 2009 at 2:40 am
Anders, it sounds like you are saying that if an organisation gets themselves a bad outsourcing contract then they can have problems. Sounds logical to me...
Original author: https://github.com/SQL-FineBuild/Common/wiki/ 1-click install and best practice configuration of SQL Server 2019, 2017 2016, 2014, 2012, 2008 R2, 2008 and 2005.
When I give food to the poor they call me a saint. When I ask why they are poor they call me a communist - Archbishop Hélder Câmara
February 6, 2009 at 8:24 am
Whenever you have a sourcing contract, you need to be sure you have this spelled out with the provider. If it's a small provider, you should also think about escrowing code as well.
Viewing 11 posts - 1 through 10 (of 10 total)
You must be logged in to reply to this topic. Login to reply