|
ugh.
What I don't like is it switches to the error panel whenever anything important happens, but that doesn't show you output from pre-build steps very well so i find myself constantly clicking back to output after compile.
Real programmers use butterflies
|
|
|
|
|
I hate the error panel. Problems are listed in more-or-less random order. At least with the Output window I have a chance of identifying the real problem.
Software Zen: delete this;
|
|
|
|
|
But the output in the Error List view is sortable. Just click on one of the column headers and it will sort into the order of data in the specified column. I've only goy VS2015 and VS2017; so it might be something that is no longer available in VS2019. If they have dropped it, remember the MS design philosophy "Redmond knows best".
Of course, my coding is always perfect, so the Error List is always empty 
|
|
|
|
|
So at my job, I have 2 business analysts (BA's) that write up all the stories for us to work on. One of them is a power user with industry knowledge and has worked in Support on the program for years, but no computer classes. The other has a degree in Computer Science, but no industry knowledge. The rules of my office are that the BA's call the shots for anything UI related.
I have more problems with the second BA (the one with the degree) designing bad UI's. So bad that my boss's boss told us to fix one part because it was so bad. This BA constantly ignores industry standard UI design principles and doesn't follow our program's look and feel when new parts are developed. They also always want to do things in a "new and better" way. This, I believe, breaks the UI contract we have developed over 20 years with our customers. It makes the program a hodge-podge of styles/designs, so the customer never knows how it's going to work from one area to the next.
As for the first BA, I get along great with. She gets it 100%.
Any advice?
Bond
Keep all things a simple as possible, but no simpler. -said someone, somewhere
|
|
|
|
|
Matt Bond said: The rules of my office are that the BA's call the shots for anything UI related.
and: so bad that my boss's boss told us to fix one part because it was so bad
Easy … get your boss's boss to change the rules of the office! 
Seriously though - either break the rules yourself or make it your boss's boss's problem.
|
|
|
|
|
First, try to talk to the BA.
If he doesn't listen you may want to escalate things with your boss.
Whatever you do, make very sure people know who does the UI design.
If your boss's boss tells you to fix something and your boss knows who did the initial design, chances are he's going to talk to the BA for you.
|
|
|
|
|
If the boss and boss's boss don't realize the problem or are unwilling to fix it, there's really not much you can do except tediously document each issue and present your documentation at some point to the boss or boss's boss.
|
|
|
|
|
Matt Bond wrote: constantly ignores industry standard UI design principles
Are the design principles specific to your industry or design in general? I ask because if your program's UI is 20 years old, maybe updating things to be easier to use isn't a bad thing.
I agree with Sander that you should try talking with the BA. Explain why the design needs to be consistent and if there are industry specific standards to follow, those as well. At the same time, be open to suggestions that might improve usage. Maybe some of the "new and better" ideas might improve things for the end users... would it be an improvement to get rid of wasted clicks even if they are 'standard'?
That said, a lot of the CompSci people I've worked with (myself included) are terrible at UI. We don't always think about how things should work the same as 'normal' people do. It wasn't until I spent a few years with actual trained usability experts that I started to change my approach to something akin to 'If you have to think about it then it probably isn't a good UI.'
|
|
|
|
|
I don't have much advice other than don't hire people without industry experience - something I learned in the .com boom days.
Academia doesn't teach proper software development, and because academia is sort of antithetical in nature to putting practice over theory it probably never will.
People get lost in the Right Way(TM) to design things that often isn't the right way at all when the real world collides with it, but they get stuck in this idea of "it must fit the model" and i've found that the higher the degree in Compsci the more likely this sort of thing comes from them.
I'm sorry to say but uni as much as it educates, puts blinders on people, and those blinders only get undone through actual experience.
If you want to save this hire, be patient with them, correct them, and hope to hell they are responsive to criticism and can update their work accordingly.
Otherwise, it's time to look for a different UI person. Just my $0.02
Real programmers use butterflies
|
|
|
|
|
I couldn't agree more. Just yesterday I was looking at some code by our newest hire and I had to correct several things they did because that was the way they were taught in school. I had to tell them that was fine for small, assignment projects but for larger scale, real world projects those are definitely not the right ways to things.
"They have a consciousness, they have a life, they have a soul! Damn you! Let the rabbits wear glasses! Save our brothers! Can I get an amen?"
|
|
|
|
|
A company BA you're likely stuck with. A contractor (you didn't hire) you can shuffle aside.
The whole MVVM mentality totally divorces the (BA's) view from the physical and logical data models and any performance considerations.
Share your models. If you don't have any, start making them (Visio). Then design and code.
It was only in wine that he laid down no limit for himself, but he did not allow himself to be confused by it.
― Confucian Analects: Rules of Confucius about his food
modified 23-Jan-20 11:00am.
|
|
|
|
|
Matt Bond wrote: I have 2 business analysts (BA's)
Not advice, but a reminder that you can't spell "analyst" without "anal"...
|
|
|
|
|
I really hate to say this, but what you need is to have regular meetings about design.
Sometimes, meetings do actually serve a good purpose.
I wanna be a eunuchs developer! Pass me a bread knife!
|
|
|
|
|
Write a design guide, put a "printed" date on each page of [x] years before the new BA started, give it to the dog to sleep on for a few nights, then refer to it whenever he gives you work. Just point out it's non-conformant. When he asks why no-one's produced it before, just say it's so internalised that it's second nature to everyone. But must be adhered to.
|
|
|
|
|
Find your cousin James, and ask him to terminate the 2nd BA with extreme prejudice.
Software Zen: delete this;
|
|
|
|
|
I am trying to use Robocopy switches.
I have a drive that is mapped as \\sr\archive
I have a local drive s:\AZ\A
this code is the normal command :
Robocopy \\sr\archive s:\AZ\A /MIR /FFT /Z /XA:H /W:5
Except I need to have a way to keep things up to date.
Do I put something in Task Scheduler, or can I use the /mot:5
switch and will it just keep Robocopy in memory that way?
how can I robocopy 7 paths which have a file name over 300 character ?
|
|
|
|
|
If you remove all the folders on the source it'll not only work, but be much faster too.
better yet, try FORMAT - can't think of it right now but there's a switch that'll make it work.
after many otherwise intelligent sounding suggestions that achieved nothing the nice folks at Technet said the only solution was to low level format my hard disk then reinstall my signature. Sadly, this still didn't fix the issue!
|
|
|
|
|
"/C" always did it for me...
"I have no idea what I did, but I'm taking full credit for it." - ThisOldTony
AntiTwitter: @DalekDave is now a follower!
|
|
|
|
|
|
I strongly suggest that you look up exactly what the FORMAT command does, followed by this link: Cambridge English Dictionary[^]
"I have no idea what I did, but I'm taking full credit for it." - ThisOldTony
AntiTwitter: @DalekDave is now a follower!
|
|
|
|
|
Ask the people who created it: robocopy | Microsoft Docs[^] and certainly don't ask it where the page expressly asks you not to, in red writing ...
"I have no idea what I did, but I'm taking full credit for it." - ThisOldTony
AntiTwitter: @DalekDave is now a follower!
|
|
|
|
|
I have always used xcopy with the /d for anything with a newer date than the target.
xcopy src: tar: /c/h/e/r/y/d
YMMV Some like Robocopy, I have found it to not have something I want. I have used it in some instances where xcopy just won't work though.
To err is human to really mess up you need a computer
|
|
|
|
|
rnbergren wrote: I have always used xcopy with the /d for anything with a newer date than the target.
If that's the goal, then this is how I've always used robocopy:
robocopy source target /MIR /R:0
Even though my backup set is measured in terabytes, the above only copies what's actually changed at the source (and deletes what's been removed), and only takes a few minutes. That's been reliable for me and that's what I've been using for nearly a decade.
|
|
|
|
|
thanks dandy72, it works and about the longpath name ?
|
|
|
|
|
Good question.
Long path names are either handled correctly already by robocopy, or I've just been very lucky and have never encountered them. Must be the first one, because when it comes to this sort of thing, luck is rarely on my side.
Quick Google search says robocopy supports long paths.
|
|
|
|
General 
News 
Suggestion 
Question 
Bug 
Answer 
Joke 
Praise 
Rant 
Admin 
Submit an article or tip
