|
charlieg wrote: a 40 yo manufacturing system
charlieg wrote: reporting system: any reporting system is better than they have
They already used Crystal Reports 40 years ago? 
It's an OO world.
public class SanderRossel : Lazy<Person>
{
public void DoWork()
{
throw new NotSupportedException();
}
}
|
|
|
|
|
haha - no. years ago we would have custom s/w that generated the reports. We do this now in "report systems" that work on a different abstraction layer.
Charlie Gilley
<italic>Stuck in a dysfunctional matrix from which I must escape...
"Where liberty dwells, there is my country." B. Franklin, 1783
“They who can give up essential liberty to obtain a little temporary safety deserve neither liberty nor safety.” BF, 1759
|
|
|
|
|
I would second Guy's recommendation. However I would use a WCF to service the data requirements and probably an OData or Json transport format. This should allow you to have 1 service and any UI can feed of it.
While WPF is going to be a challenge it is well worth the effort. SQL Server and Reporting Services is really a no brainer if you have experience with them.
Look into a set of UI controls, we use Telerik and are reasonably satisfied with them. There are also alternative reporting tools but I like SSRS for the service (RDL) and embedded (RDLC) formats.
Never underestimate the power of human stupidity
RAH
|
|
|
|
|
WCF - that qualifies as shiny.
Charlie Gilley
<italic>Stuck in a dysfunctional matrix from which I must escape...
"Where liberty dwells, there is my country." B. Franklin, 1783
“They who can give up essential liberty to obtain a little temporary safety deserve neither liberty nor safety.” BF, 1759
|
|
|
|
|
charlieg wrote: WCF - that qualifies as shiny.
No, that qualifies as a solid and best practice way to access your data layer. That way all your apps can use the service to access the data; wpf, internet, mobile, etc....
Just saying.
Edit: Didn't want to come off harsh. Just want you to know, that WCF is not shiny technology. Who is going to be developing this new system, in house software engineers or are you going to have to hire out?
|
|
|
|
|
Well you do need a service layer and it must be agnostic to all client platforms and something as solid as WCF (we have been using it for 5 years and we are NOT early adopters) should be in your primary selection set. I think the shiny has been rubbed off by now.
Never underestimate the power of human stupidity
RAH
|
|
|
|
|
gentlemen - shiny means new to me 
Charlie Gilley
<italic>Stuck in a dysfunctional matrix from which I must escape...
"Where liberty dwells, there is my country." B. Franklin, 1783
“They who can give up essential liberty to obtain a little temporary safety deserve neither liberty nor safety.” BF, 1759
|
|
|
|
|
WinForm, and because of my familiarity with DevExpress, I'd use them, but I SyncFusion's stuff seems pretty good too, from what little I've done with their controls.
Since I've chosen to skip WPF (though I have had a glancing blow with one client) my personal conundrum is shifting development entirely to the web, and all the bad and ugly that comes with it. Some of the stuff I've seen is really good, but gawd, it looks painful to really learn.
Marc
|
|
|
|
|
Like Marc, I would go for the DevExpress toolkit + WinForms.
WinForms is a mature platform and the ecosystem is big enough by far.
The DevExpress WinForms controls are wonderfully integrated into WinForms.
Aside all their controls, the DevExpress reporting tools are extremely flexible, it is a dream for a developer.
Now comes the best: Do you want to be finished with your project before lunch? Then take a look at DevExpress XAF (eXpress App Framework): You will get a perfectly designed WinForms app PLUS a perfectly looking Web App, both running before lunch.
Check it out for yourself:
Amanda explaining in 10 minutes how to build an app with DevExpress XAF[^]
|
|
|
|
|
I also would have to agree with Guy. One thing to remember is the size and loading of the system. If it's small then you'd be sweet with Guy recommendation. If it's in the large size then you'll need to start thinking of breaking it down to the various layers (both database and application). Have fun, I'm a bit jealous really..
|
|
|
|
|
The good news is that just about anything will be faster than what they have. I've cobbled together the system over the years, but the one thing I don't really know is where the corporation is going. There are a lot of different groups that have their oars into the production area, so the more flexible the design, the better off I'll be.
It's actually exciting. I've been so long in coding/debugging for the past couple of years, being able to organize a design like this is refreshing.
Charlie Gilley
<italic>Stuck in a dysfunctional matrix from which I must escape...
"Where liberty dwells, there is my country." B. Franklin, 1783
“They who can give up essential liberty to obtain a little temporary safety deserve neither liberty nor safety.” BF, 1759
|
|
|
|
|
I wold bite the bullet now and do the front end in HTML5 / Javascript using something like KnockoutJS or angularJS
Your back end would be essentially the same (I concur with the SQL Server / SSRS) but provide web services to provide / update data to the client.
You then have maximum flexibility for devices to run it, have fun learning it and, using web services, still have the flexibility to use any other client too, should the fancy take you.
PooperPig - Coming Soon
|
|
|
|
|
thanks maxx, more tech to familiarize myself with. The good part is that I'll bet I can find people who can write this stuff. Ever try to find a FORTRAN programmer?
Note: I know some of you are out there 
Charlie Gilley
<italic>Stuck in a dysfunctional matrix from which I must escape...
"Where liberty dwells, there is my country." B. Franklin, 1783
“They who can give up essential liberty to obtain a little temporary safety deserve neither liberty nor safety.” BF, 1759
|
|
|
|
|
Hi,
You pretty much nailed down the core technology set. Following is just some design consideration and best-practices. As a general rule, try to minimize technology surface as much as possible. Pertinent to your case it seems that the following technology set may suffice the goal;
1.1. Backend DB - Sql Server Compact Edition 4.0 (single-file based DB: easy to deploy/backup/distribute)
1.2. Business Layer - C# 4.0 (Client Profile redistributable)
1.3. Front End - WPF/XAML (.NET 4.0, the same Client Profile redistributable)
1.4. Optionally - implement data export to Excel feature for 'ad-hoc' reporting/analysis
As for sample practical implementation of the aforementioned technologies, you can refer to my recently released app PaydayNY-2014 for Win 7/8 to see how it works in real life (there is a free Trial edition) PaydayNY-2014P-Pro-for-Win[^]
Best regards/wishes,
<lol>Life is 2short 2remove USB safely
|
|
|
|
|
Html5/JS or one of the frameworks that make it easier to develop browser based UI. Aside from cross platform cross device advantage current Microsoft UI solutions can have limited support in the future or be deprecated since Microsoft is in transition now with no clear future direction for old UI stack. You may also check other DB solutions more modern than SQL if they are more suited to solve your problem (along the lines of document DB (Mongo, Hadoop etc.) and not relational DB).
|
|
|
|
|
First of all I would approach the "conundrum" of designing a UI that is "flexible" by interviewing in-depth the significant end-users of the current system. I'd listen most carefully to their descriptions of what's easy to use, and what's tricky to use, and what's damn difficult to use with the current system. I'd ask them for their ideas, doing my best to "set them at ease" ... by building trust ... so they could come up with anything they imagined and express it. I'd do the same thing with the managers of the front-line users.
I would then create sketches incorporating the end-user/manager feedback, ideas, and my own inspirations; I'd go back and show those sketches to the same folks I interviews and study their reactions.
After #n iterations of the design process, I would then attempt to formalize the specifications, and constraints, on the UI, taking into account hardware, software, security, and "organizational culture."
It may be obvious, but let me say it anyway: by involving others in the design phase you both build "mind-share" and positive expectations that may make acceptance of your final design easier, but you also off-load the risk of coming up with something that's a total surprise to the end-users, and which they may have real trouble accepting.
Only then, would I evaluate whether to use WinForms, WPF, or the Web Stack, to implement the UI, based on criteria I'd developed in the design phase: need for scaling while preserving look-and-feel across any range of sizes (I'd go with WPF for that); need for whiz-bang graphic effects and animation (again, WPF).
All this should not constrain your creative imagination: if you come up with a powerful "vision" of a radical new design that might have a steep learning-curve for current users. Well, so then you have a sales job to do with both end-users and their managers. I would certainly not start that sales job until I had a smoothly working prototype, and estimation of training costs, and eventual benefits of adoption.
If the various musings above are not relevant at all to your context ... sorry 
« I had therefore to remove knowledge, in order to make room for belief » Immanuel Kant
|
|
|
|
|
You suggestion is just perfect in an ideal world...Once I had the opportunity to build some software from ground-up and went to do the interviews...
After 10 weeks (and it was within time-frame) I had a UI design confirmed by all relevant person - at the next week they announced bankruptcy...
I never ever since then had the opportunity to do the same...
I'm not questioning your powers of observation; I'm merely remarking upon the paradox of asking a masked man who he is. (V)
|
|
|
|
|
While I am addicted to imagining I am free of constraints, in spite of overwhelming evidence to the contrary ... in this case I would assert that the method you pursued ... interviewing, design phase iteration ... like the method I advocate ...
Is "real world" !
What is "ideal world" is for a programmer/designer to ignore existing "context" ... current software and its users and managers, company culture, existing hardware capabilities, learning curve for new software, etc., ... and imagine they have a tabula rasa (blank slate) on which they can create any damn thing they imagine.
Do you really think your 10 weeks of design-phase was a cause of the company's bankruptcy ?
« I had therefore to remove knowledge, in order to make room for belief » Immanuel Kant
modified 28-Sep-14 5:25am.
|
|
|
|
|
Love the way he jumps straight to various Windoze technologies. Yeah. Love that.
|
|
|
|
|
Bill,
A good point, but what they don't know is that they really need to replace this system. As I'm wrapping up some very late changes for them, partly because of how old this system is, partly because of my mgt. skills, this is somewhat of a ninja project.
If I can get a system basically running without too much public involvement, then I'm halfway home. True, I may toss it all in the bin, but I have the hooks into the people that really matter - the end users. It's interesting. The mgt. team has tried to force feed a new system into the factory and production process, and it failed miserably for the exact reasons you mention - they failed to ask the end users what they need.
At the end of the day, I'll have learned a lot of tech.
cg
Charlie Gilley
<italic>Stuck in a dysfunctional matrix from which I must escape...
"Where liberty dwells, there is my country." B. Franklin, 1783
“They who can give up essential liberty to obtain a little temporary safety deserve neither liberty nor safety.” BF, 1759
|
|
|
|
|
Be very careful with the interviews. You will get scope creep and then everyone will be expecting something different.
|
|
|
|
|
Before you make any decisions, you should read what Joel says[^] about rewriting software.
|
|
|
|
|
What an excellent point... I need to go get some coffee and ponder this....
Charlie Gilley
<italic>Stuck in a dysfunctional matrix from which I must escape...
"Where liberty dwells, there is my country." B. Franklin, 1783
“They who can give up essential liberty to obtain a little temporary safety deserve neither liberty nor safety.” BF, 1759
|
|
|
|
|
That said, there are perfectly valid reasons to start over from scratch.
For example, if the legacy code is done in lisp.
Anyway, depending on all those thing I don't know about your project, I would probably refactor the code into logical modules/ layers that can be easier exchanged or expanded to newer technologies, when needed, one by one.
So start with planning the architecture you want, and then how to get there.
|
|
|
|
|
I cannot count the number of times I have linked to that article in the past 10 years. This was pure gold advice !
~RaGE();
I think words like 'destiny' are a way of trying to find order where none exists. - Christian Graus
Entropy isn't what it used to.
|
|
|
|
General 
News 
Suggestion 
Question 
Bug 
Answer 
Joke 
Praise 
Rant 
Admin 
Submit an article or tip
