[Info-vax] Java on VMS, was: Re: So is there still a hobbyist program or not

[Arne Vajhøj]

On 7/20/2019 6:16 PM, Scott Dorsey wrote:
> =?UTF-8?Q?Arne_Vajh=c3=b8j?=  <arne at vajhoej.dk> wrote:
>> I think it is rare that applications get totally rewritten often. Cost
>> prevents that.
>>
>> Drivers for rewrites are usually one or a combination off:
>> 1) the platform is burning aka no support from vendor and
>>     difficult to find people with skills
>> 2) changes in general IT strategy where the platform is
>>     no longer an option because it will not run in the new
>>     environment
>> 3) there is a business need for very significant enhancements
>>     that just are not practically possible with current
>>     platform
>> 4) there is a significant cost saving by changing platform
> 
> The aforementioned Lubri-Moly issue is a case of #4, sort of.  The idea was
> that by moving from proprietary applications designed specifically for that
> business to commercial general-purpose business applications that there would
> be cost savings in software maintenance, and that the software would in the
> end be more flexible.
> 
> This is not the case, as anyone who has ever seen an SAP disaster can attest.

There is typical not anything wrong with the logic when the project
starts.

A standard application is much lower cost to maintain than a custom
application.

*IF* it really is standard.

The problem is that it often goes like:
* pick a standard application that will ensure low cost
* ask every person with a fancy title what the requirements are
* they provide tens of thousands of requirements usually
   to provide the same functionality as the old custom
   application
* the standard application get heavily customized
* the initial delivery get NN months delayed and goes
   XXX M$ over budget
* whenever the standard application get a great new feature
   that they want then merging the new standard stuff
   with the customizations cost a small fortune

Arne