April 12, 2016 at 8:10 pm
Comments posted to this topic are about the item Build a POC
April 13, 2016 at 12:06 am
True, POC's are usefull.
But most of the companies today don't give that much time to developers to try something new or an alternative approach and come up with POC's.
They follow the old patterns and call themselves innovative 🙂
April 13, 2016 at 12:43 am
Very recently I have seen a team (including management) use the term POC when what they were doing was Evolutionary Prototype i.e. there was no intention to throw away the code unless it was an unmitigated disaster. I have yet to view the code and I know the team are very capable but pretending that the initial development was a throwaway POC then changing mindset to it being production code (eventually) is extremely dangerous. I am betting that some of the early code is not as maintainable as the latter code.
POCs are great...but if you build it someone else may ship it. So you need to know whether you are truly building a Throwaway Prototype or are actually developing an Evolutionary Prototype.
Gaz
-- Stop your grinnin' and drop your linen...they're everywhere!!!
April 13, 2016 at 2:26 am
I've lost count the number of times a POC or demo set up becomes the production instance after sales and account managers realize they can just start billing clients straight away. They're then surprised when capacity and stability issues start within a couple of weeks/months...
April 13, 2016 at 7:28 am
Would a Concept model be a safer first step? Once reviewed by interested parties a POC could be initiated on a sandbox test server to give the concept a good workout. A successful result could then go through the normal logical and physical modeling prior to becoming a real project destined for production when it is ready.
April 13, 2016 at 7:28 am
Gary Varga (4/13/2016)
... there was no intention to throw away the code unless it was an unmitigated disaster.
That sounds like what we do here.
April 13, 2016 at 7:47 am
POCs aren't the issue. The main issue is management and developers not wanting to throw away code. The temptation to reuse something that's mocked or just setup is great. However you should be able to rebuild the code, in a better way, in a fraction of the time of the first one. Plus, it's a chance to redo things that weren't done well.
Can't help that. I don't think avoiding a POC is the way to do this, however.
April 13, 2016 at 8:20 am
Steve Jones - SSC Editor (4/13/2016)
POCs aren't the issue. The main issue is management and developers not wanting to throw away code. The temptation to reuse something that's mocked or just setup is great. However you should be able to rebuild the code, in a better way, in a fraction of the time of the first one. Plus, it's a chance to redo things that weren't done well.Can't help that. I don't think avoiding a POC is the way to do this, however.
Part of the problem is developers who blur the line between PoC and beta/alpha version of a product. Some of the code from a PoC might and probably should be reusable but it shouldn't be presented as something that can be deployed in even a bare bones minimally functional way.
April 13, 2016 at 8:32 am
Completely agree. This is one area developers' laziness is an issue.
Personally, it should be a challenge to rebuild quickly, using knowledge, but correcting the the things you noticed weren't well done in the POC.
April 13, 2016 at 9:50 am
Ah, the joy of overloaded initialisms...
In this case, if you don't throw away the POC you end up with a system that's a POC?
Thomas Rushton
blog: https://thelonedba.wordpress.com
April 13, 2016 at 10:05 am
Steve Jones - SSC Editor (4/13/2016)
...I don't think avoiding a POC is the way to do this, however.
Totally agree.
Gaz
-- Stop your grinnin' and drop your linen...they're everywhere!!!
April 13, 2016 at 10:12 am
Steve Jones - SSC Editor (4/13/2016)
...This is one area developers' laziness is an issue...
I have to take umbrage with this. Almost every example I can think of this occurring is down to pressure from management (usually above project managers too). "JFDI" / "It works. Ship it!!!"
The one time when it wasn't was when a developer with kudos but near zero education and experience refused to listen to others experience and knowledge of the inherent problem of treating a Throwaway Prototype as an Evolutionary Prototype.
Gaz
-- Stop your grinnin' and drop your linen...they're everywhere!!!
April 13, 2016 at 10:21 am
Gary Varga (4/13/2016)
Steve Jones - SSC Editor (4/13/2016)
...This is one area developers' laziness is an issue...I have to take umbrage with this. Almost every example I can think of this occurring is down to pressure from management (usually above project managers too). "JFDI" / "It works. Ship it!!!"
The one time when it wasn't was when a developer with kudos but near zero education and experience refused to listen to others experience and knowledge of the inherent problem of treating a Throwaway Prototype as an Evolutionary Prototype.
You can, but I see it both ways. I see developers thinking they can save time by reusing code that isn't written well. I've done it myself. Not proud of it, but sometimes it takes pressure off the immediate needs.
April 13, 2016 at 10:57 am
Hey Steve!
Why don't you like AdventureWorks?
April 13, 2016 at 11:26 am
Steve Jones - SSC Editor (4/13/2016)
Gary Varga (4/13/2016)
Steve Jones - SSC Editor (4/13/2016)
...This is one area developers' laziness is an issue...I have to take umbrage with this. Almost every example I can think of this occurring is down to pressure from management (usually above project managers too). "JFDI" / "It works. Ship it!!!"
The one time when it wasn't was when a developer with kudos but near zero education and experience refused to listen to others experience and knowledge of the inherent problem of treating a Throwaway Prototype as an Evolutionary Prototype.
You can, but I see it both ways. I see developers thinking they can save time by reusing code that isn't written well. I've done it myself. Not proud of it, but sometimes it takes pressure off the immediate needs.
Fair enough. Your experience (of both yourself and others) is valid, of course. It just hasn't matched mine.
Gaz
-- Stop your grinnin' and drop your linen...they're everywhere!!!
Viewing 15 posts - 1 through 15 (of 16 total)
You must be logged in to reply to this topic. Login to reply