Aging Software

  • Comments posted to this topic are about the item Aging Software

  • I feel it is perfectly reasonable for a vendor to communicate the lifecycle of their software. Lets face it Microsoft give ample warning and a long period of time when their software will be supported, enter extended support and then have withdrawn support. Case in point Active Directory 2003 which went out of support in February 2015.

    If I was running a small company I would worry about the overhead of supporting aging versions of my software. I simply wouldn't have the manpower. Perhaps I would offer customers discounts and incentives to upgrade however the discount game is very dangerous for small players where margins are tight.

    One horror story I was told was where a company had been paying for critical support only to have an incident which required that support to be enacted. It became horribly apparent that the support provider had gambled on the stability and maturity of the system and had not retained the skills to meet their contractual obligations.

  • I used to work for a hospital that had a clinical letters system written in Clipper, for an XBase file based database, that was 15-20 years old when replaced with a modern web based solution. There was a lot of resistance from the secretaries who liked the keyboard shortcuts of the old system, and feared that a new interface would slow down their work rate.

    The problem was after that many years we only had three members of staff who could program Clipper well enough to provide support, and they were the IT Director, Head of IT and Head of Information. Which would mean that upgrades or bug fixes would cost too much, and be nearly impossible to schedule. I'm not even sure that the compiler would install on a modern OS.

    There comes a time when, no matter how popular a piece of software, that you have to replace it.

  • I know one company that was still "supporting" their application running against SQL2005 until it was kindly pointed out that Microsoft had stopped supporting any & all issue with it. Now they "supporting" their application against SQL2008 , ho hum.

    To some extend isn't the industry going the other way as well. How many new languages & frameworks are being used year (and how many don't make it to market or drop into extinction). So instead of being concerned about keeping the skillset relevant for supporting old/legacy systems we now have to be aware of the all there new ways of doing things.

  • I'm faced with a personal software upgrade issue now. Microsoft would love for me to upgrade to Windows 10 on my laptop. Unfortunately Windows 10 doesn't support certain older versions of SQL Server and other software I need to have available. So I may have to maintain two separate Windows OS instances to support everything I need.

    And that's the point. It's the needs of the user/client that are often important. Sometimes newer software is a better choice, but not everyone can afford the license costs, the testing, the recertification of software, just to have a database that doesn't use any of the new features from the latest 3 versions.

    As for developing with new technology, you have to drag the whole infrastructure along. Any tools, documentation and utilities may have to be modernized or replaced. And you lose a good percentage of your investment in the older technology. There's a reason that there's still systems running Cobol, Fortran and tech of that ilk, it's just too costly to reinvent the wheel.

  • Alex Gay (10/7/2015)


    The problem was after that many years we only had three members of staff who could program Clipper well enough to provide support, and they were the IT Director, Head of IT and Head of Information. Which would mean that upgrades or bug fixes would cost too much, and be nearly impossible to schedule. I'm not even sure that the compiler would install on a modern OS.

    There comes a time when, no matter how popular a piece of software, that you have to replace it.

    True. But whether early replacement or later, there always seems to be issues. The improvements are new so additional training is needed. There are things that don't work as well (or don't exist at all) in the new code that worked in the old so animosity exists. The best adopters are the new employees and those you have chosen as your "Champions" for the new software, those that you have allowed to participate from beginning to end with the development. As to operating system upgrades, we do them as seldom as possible to avoid the required rewrite of some code. Yes, we're still on 2008 R2. Maybe I'll be retired before we switch.

  • Alex Gay (10/7/2015)


    ...

    The problem was after that many years we only had three members of staff who could program Clipper well enough to provide support, and they were the IT Director, Head of IT and Head of Information.

    ...

    I'm glad to hear that the XBase / Clipper developers had an opportunity to move on to bigger and better things within the organization while performing only occasional support on the system. That's actually a healthy lifecycle.

    In some organization is not just aging software, but aging software developers who are out of date and need upgrading.

    "Do not seek to follow in the footsteps of the wise. Instead, seek what they sought." - Matsuo Basho

  • My problem with "new and improved" software is that the software vendor frequently decides to no longer support features that are actually incredibly useful. To wit, it may be "new" but it's not always "improved" for anyone except the software vendor.

    A great example of this is the newer date/time datatypes they came out with in 2008, which have much less functionality than DATETIME. If they ever decide to get rid of the DATETIME data-type, then I will make a pork-chop run on Redmond. 😉 And deciding to no longer support the very old sp_MakeWeb task was a real killer for me when than little goodie was first announced. SSIS isn't the answer to everything! :sick:

    --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)

  • There's a lot of factors to weigh in: does the vendor still exist, do they still have the source code, etc. Do we still have people running IE 6 because of OCX controls because software couldn't be upgrade and have it still work on newer, more secure, platforms? Perhaps the better question would be 'for how much longer will we have people running IE 6 because of OCX controls'. They're going away, but oh so slowly, because management doesn't see them as a problem until something horrible happens.

    An internal system running SQL 6.5 for door access? That can run for next to forever as long as the door access hardware continues to be accessible. And I think that's cool that you have a fully-depreciated system that's still productive. All it costs is the electricity the server eats and a little time to keep half an eye on backups and DBCCs.

    Running old platforms can be fine IF they are isolated from the internet, otherwise they're a security disaster waiting to happen.

    -----
    [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]

  • Ultimately I'm mostly OK with the current way most vendors support software. If it works for a decade and support ends, I can continue to use it. Until it doesn't work, and then I am glad that most vendors have an upgrade for me.

    One thing I admire about the Wizards of Armonk (aka IBM) is their ability to keep application software running on their successive operating systems and RDBMSs. Microsoft, by comparison, is a PITA upon which to build durable application software.

    Microsoft has some advantages, but it's about the last platform I'd chose if I had any interest in preserving an investment in application software.

  • Alex Gay (10/7/2015)


    I used to work for a hospital that had a clinical letters system written in Clipper, for an XBase file based database, that was 15-20 years old when replaced with a modern web based solution. There was a lot of resistance from the secretaries who liked the keyboard shortcuts of the old system, and feared that a new interface would slow down their work rate.

    The problem was after that many years we only had three members of staff who could program Clipper well enough to provide support, and they were the IT Director, Head of IT and Head of Information. Which would mean that upgrades or bug fixes would cost too much, and be nearly impossible to schedule. I'm not even sure that the compiler would install on a modern OS.

    There comes a time when, no matter how popular a piece of software, that you have to replace it.

    Ahhh, Clipper. I made a nice living for a few years with this. Maybe I should contact them. They have some $$$$? :w00t:

  • Jeff Moden (10/7/2015)


    My problem with "new and improved" software is that the software vendor frequently decides to no longer support features that are actually incredibly useful. To wit, it may be "new" but it's not always "improved" for anyone except the software vendor.

    A great example of this is the newer date/time datatypes they came out with in 2008, which has much less functionality than DATETIME. If they ever decide to get rid of the DATETIME data-type, then I will make a pork-chop run on Redmond. 😉 And deciding to no longer support the very old sp_MakeWeb task was a real killer for me when than little goodie was first announced. SSIS isn't the answer to everything! :sick:

    Agree on the first point. I sometimes hate the changes that remove old stuff I use.

    On datetime, that's never leaving. It's in so much code, that I can't ever see this being removed.

  • Several years ago I did some consulting for a client that was operating very antiquated software, i.e., Visual Basic 4.0 and Sql Server 6.5. on old server platforms. They were operating on NT4.0 and were purchased in the late 1990's. The client would not purchase new equipment or upgrade their software. It was a difficult client to work with. As a result of the bad experience with these folks I decided I would never again work with a client that operated antique equipment or software.

  • ... I decided I would never again work with a client that operated antique equipment ...

    If it's possible to get parts for a '63 Studebaker, it ought to be possible to keep old systems running. (Then again, Studebaker is long gone. And Microsoft isn't ... at least not yet.)

  • Some commentators have already pointed out that as we move into IoT, some of the things on the internet will have working lives measured in decades and many will have no online interface to allow a software upgrade. We already see many industrial control systems that are 20 or more years old with no plans for replacement.

    This puts a completely different spin on software support. For some items we can take the attitude that while the thing works we continue to use it (and hope it has not been hacked to spy on us) and when it dies we look for something else. When the thing is the switch that controls the electricity or water supply to your part of town then longevity of service and support becomes important.

    There will remain some parts of the IT industry that thrive on ephemera, such as smartphones and other devices that are very regularly updated. However, I think that the majority of the IT industry will become increasing tied into supporting long lifecycle code. This will apply to both system software such as SQL Server and applications. Most vendors should be happy about supporting a much longer lifecycle providing that it can give them an income stream.

    IoT is seen as disruptive for many aspects of IT. It will be interesting to see in 30 years time how it has affected software support.

    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

Viewing 15 posts - 1 through 15 (of 54 total)

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