We are a Microsoft shop




  • I was in a "5-year Technology Landscape" meeting today where a client decided to use SQL Server instead of Oracle because they are a "Microsoft shop". Fair enough. Then they also decided to use VMware instead of Microsoft Hyper-V because VMware is better and they plan to switch internal development to Java because it is more stable than C#. TFS? Not robust enough, let's bring in Rational. And Tivoli for backup and OpenView for monitoring (after all, who cares if the servers are IBM, the PC are HP machines). Email? Apparently Domino works better with Blackberry. SharePoint sucks, long live Alfresco. Nobody disputed using Windows and MS-Office on desktops, but of course application virtualization has to be done in Citrix.

    Larry Ellison might have his own island and a jet, but his product will never set foot in this "Microsoft shop".



  • One may walk the road to enterprisey, but without Horricle and a liberal sprinkling of XML, you'll never reach your destination.

    - Confusedus, 49BC



  • Java is more stable than C#? Do they just mean that C# changes too damn much?



    TFS is not robust? That's news to me.



    I read the Wikipedia page on OpenView and I still can't discern how it is useful. If your people try and do streaming video or use torrents, your network will be slow. If they don't, it will not. You don't need Neo to run a network, just get someone who's a pushover and won't get in the way, and don't let him/it spend any money unless it's directed at an obvious, in-your-face failure.



    SharePoint does suck, but looking for some other version of SharePoint that doesn't suck is like looking for the "good" Holocaust.



    Citrix is a slow, flaky piece of garbage foisted upon us by the lazy / paranoid.



    I'm not a huge fan of Oracle any more, though. Postgresql basically does everything I would use Oracle for now. People need to get up-to-speed on Postgresql. This is not 2000... no one should be writing huge checks to Oracle / Microsoft just because they want an RDBMS.



  • @bridget99 said:

    SharePoint does suck, but looking for some other version of SharePoint that doesn't suck is like looking for the "good" Holocaust.
     

    That's the 9/11 of metaphors.



  • @bridget99 said:

    Java is more stable than C#? Do they just mean that C# changes too damn much?



    TFS is not robust? That's news to me.



    I read the Wikipedia page on OpenView and I still can't discern how it is useful. If your people try and do streaming video or use torrents, your network will be slow. If they don't, it will not. You don't need Neo to run a network, just get someone who's a pushover and won't get in the way, and don't let him/it spend any money unless it's directed at an obvious, in-your-face failure.



    SharePoint does suck, but looking for some other version of SharePoint that doesn't suck is like looking for the "good" Holocaust.



    Citrix is a slow, flaky piece of garbage foisted upon us by the lazy / paranoid.



    I'm not a huge fan of Oracle any more, though. Postgresql basically does everything I would use Oracle for now. People need to get up-to-speed on Postgresql. This is not 2000... no one should be writing huge checks to Oracle / Microsoft just because they want an RDBMS.

    I was reporting what the client said. I could go and challenge their preconceived ideas but whatever the technology, they will probably do a poor job at implementing it anyways.



  • @Speakerphone Dude said:

    switch internal development to Java because it is more stable than C#.
    They're going to move from C# to Java and their developers aren't rioting yet? 



  • @TGV said:

    @bridget99 said:

    SharePoint does suck, but looking for some other version of SharePoint that doesn't suck is like looking for the "good" Holocaust.
     

    That's the 9/11 of metaphors.

    And you're the Seth McFarlane of TDWTF.

     



  • @DOA said:

    @Speakerphone Dude said:

    switch internal development to Java because it is more stable than C#.
    They're going to move from C# to Java and their developers aren't rioting yet? 

    The internal development team is just a bunch of managers and PMOs. This client mostly relies on offshore contractors for development.They don't develop a lot of applications, they use COTS and just hire contractors to build plugins, add-ons and small additional features.

    A few years ago almost everything was in-house, business users were using homegrown apps and the development group was pretty big (50+ including DBAs, architects, etc.). I don't have all the details but from what I understand they had a legacy system running on OS/400 and IT sold the business on rewriting everything in C#, with all the bells and whistles (web services, SharePoint integration, etc.). After two years the C# version had a minimal set of features and was pretty shaky but IT wanted to pull the plug on the old system so they could focus their effort on "stabilizing" the new stuff. Instead the C-level people got pissed; the company brought in IBM and Big ERP and fired almost everyone in the IT group.

    I could tell this was a difficult time for them, as you can feel the temperature dropping in the room when someone mentions C# or web services. But from what I see, no matter how bad was the legacy or C# solutions, the current stuff they have is probably worse.

    In my opinion this was still a good call from senior management, otherwise they would not currently be financing my next trip to Vegas...



  • @bridget99 said:

    I'm not a huge fan of Oracle any more, though. Postgresql basically does everything I would use Oracle for now. People need to get up-to-speed on Postgresql. This is not 2000... no one should be writing huge checks to Oracle / Microsoft just because they want an RDBMS.
    Don't tell Morbs that.  He'll rip you a new on on how PostgreSQL sucks and MySQL is The One Path to Database Enlightenment.



  • @da Doctah said:

    @TGV said:

    @bridget99 said:

    SharePoint does suck, but looking for some other version of SharePoint that doesn't suck is like looking for the "good" Holocaust.
     

    That's the 9/11 of metaphors.

    And you're the Seth McFarlane of TDWTF.
     


    <font size="7">     ZING!!</font>



  • @bridget99 said:

    I read the Wikipedia page on OpenView and I still can't discern how it is useful. If your people try and do streaming video or use torrents, your network will be slow. If they don't, it will not. You don't need Neo to run a network, just get someone who's a pushover and won't get in the way, and don't let him/it spend any money unless it's directed at an obvious, in-your-face failure.

    I would hope that anyone reading the Wikipedia page on OpenView would mostly discern that the Wikipedia page on OpenView has virtually no useful content to anyone who is not already versed in OpenView.  The interesting bits that actually talk about what OpenView may be able to do are on the Network_management and System_management pages.  That's particularly useless, because they don't talk about what OpenView can actually do, just what products in its market space can generally do.

    The main selling point for HP OpenView is that it can be configured to perform near-real time monitoring of all of the components of your internal network, including servers, routers, printers, and desktops (note: the utility of the last bit tends to be rather low, since people frequently shut them down at night - or, in the case of laptops, take them home.  Also, if you're lucky enough to be running in a sane environment, there are no mission critical servers running on desktops (or laptops), as they generally lack data center backups, including UPS power, and people frequently shut them down at night, or in the case of laptops, take them home.  With any luck, tablets such as iPads are right out.)  It does this via a very complex and confusing interface, so the people managing that environment can charge top dollar rates.  HP OpenView works *almost* as well as several competing free/open source software packages, such as Nagios, so you lose relatively little capability in exchange for spending a lot of money on OpenView and the people who are tasked with maintaining it and watching over it.

    I think my personal favorite capability of OpenView is its ability to be configured to work entirely off of SNMP traps and pings, so all it takes is one misconfigured/overzealous firewall, and an entire section of your system works absolutely flawlessly as any of your reports to management indicate.  (Since traps only fire when a problem happens or is resolved, they're far more efficient from a performance standpoint.  However, since they depend on the potentially faulty system to identify itself as potentially faulty, many of the more security minded of us have a certain amount of reservations about their utility for monitoring security systems.  It's also rather fun to see an OpenView server keel over and die when someone shuts the data center AC off over a three day weekend, and every system starts sending off nearly every trap it has.)

    Note that if you're running a relatively small shop, having a real-time network monitoring system may not be worth the cost for several years.  Their purpose is to reduce the duration of service downtime by notifying the system administrators about it as soon as possible.  When one has hundreds of servers, even if one has very stable servers, it's usually worthwhile at least once a month.  When one only has a couple of servers, you can go for years without a network or server outage.  Then you have one, nobody finds out about it until your customers are all switching over to your competitor because their server's up, and you go out of business.  But at least you didn't waste money on stupid network monitoring for the years you were in business.

    (Disclaimer: the above small company example is just meant to be hypothetical.  It is not intended to resemble any real world small company experiences.  Especially not any of my father's small businesses.  Apart from the fact that he generally doesn't have any network monitoring, because he only has one server and one server monitoring itself is at least as silly as using SNMP traps for all of your monitoring.)

    (Disclaimer II: Nagios has its own problems.  Also, like most of the products in this market, it too can be configured to work entirely off of SNMP traps.  I just haven't seen anyone actually *do* that, and most of the OpenView implementations I've seen were configured that way.  Note that Nagios' typical performance problems can be reduced drastically by using real programming languages for its helper scripts, rather than going with the suggested Bourne Shell scripts.  Anyone familiar with how Bourne Shell scripts perform compared to many of the alternatives available in a unix environment should know this.)



  • They won't use Oracle, but are fine using Java, which is owned by Oracle?

    How am I the only one to bring this up?



  • @GrizzlyAdams said:

    They won't use Oracle, but are fine using Java, which is owned by Oracle?

    How am I the only one to bring this up?

    They even use one of Oracle's ERP... but store ERP data in SQL Server. The only DBA that survived the layoffs is Oracle certified. And I'm pretty sure Oracle sales rep offer bigger bribes incentives than Microsoft. Go figure.



  • @GrizzlyAdams said:

    They won't use Oracle, but are fine using Java, which is owned by Oracle?

    How am I the only one to bring this up?

     

    oracle purchased, but did not develop, java; they may not have heard.

     


Log in to reply
 

Looks like your connection to What the Daily WTF? was lost, please wait while we try to reconnect.