|
I come form "flow charts" and desk checking. I learned about machine hours too; except all my time was spent running test cases for QA trying to keep up with the speed of development (I wasn't "compiling", as I explained).
I used to go to the art store to buy things like "French curves" in order to make DFD diagrams for my clients. I then moved on to Visio. At some level, I can't function without some diagramming. I don't know how others manage without. They've established a level of mediocrity which has become the standard (i.e. Agile).
I did "agile": 13 planned releases over 26 weeks. Some 10 developers. Hit every milestone. Preceded it with 6 months of analysis and design though; and another 6 months for the "DBA's" to "get it" and to see that "our" DB design was also spot on.
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
|
|
|
|
|
Gerry Schmitz wrote: I did "agile": 13 planned releases over 26 weeks. Some 10 developers. Hit every milestone. Preceded it with 6 months of analysis and design though; and another 6 months for the "DBA's" to "get it" and to see that "our" DB design was also spot on. That sounds fantastic
We get a two week sprint in which to understand, design and code the solution to a user story.
We do get refinement sessions in advance for each user story, although we are not "allowed" to specify any solutions within user stories which in my opinion defeats the purpose of refining a user story.
“That which can be asserted without evidence, can be dismissed without evidence.”
― Christopher Hitchens
|
|
|
|
|
Sometimes these things require an overall architecture as in the case of a "pipeline" (revenue, goods, whatever).
Agile doesn't seem to acknowledge this. Everything is assumed to be automatically plug compatible; and no one seems particularly interested in the big picture. I don't last in situations where I can't maintain some sort of "lead" (i.e. I don't take kindly to orders, anymore).
The "fog of war" should be avoided at all cost, IMO.
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
|
|
|
|
|
Gerry Schmitz wrote: ... because you're not spending any time on analysis and design? Don't you dare coming here with archaic riverfall ideas!
We are living in an age of agility and scrum. Analysis and design has no place in a modern world.
|
|
|
|
|
The documentation. Good grief, the documentation.
Estimated time to write software: X
Actual time to write software: 2X
Time to write docs: 3X
Testing: Well, something's got to give, right?
What do you mean you don't bother with docs? Next you'll be telling me you just redirect help queries to the web!
|
|
|
|
|
I've also seen something like
- time to write software that you will use briefly: x
- time to write software that you will use for a long time: 3x
- time to write software that other people will use: 9x
|
|
|
|
|
Another, similar one:
Resources to write a program: x
Resources to write a program componment (i.e. a program that must relate to and interact with other components in a comlex environment, adhere to standards, ...): 3x
Resources to create a program product (i.e. a program with documentation, a support line, marketing & sale elements, ...): 3x
Resources to develop a program component product: 10x
I think these estimates are roughly 40 years old. I have never seen them contradicted in practice.
|
|
|
|
|
I start out with the objective that the software shouldn't need documentation (other than tool tips) if it's done right. Match the user's "mental model". So far I'm succeeding.
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
|
|
|
|
|
Gerry Schmitz wrote: the software shouldn't need documentation (other than tool tips) if it's done right. Match the user's "mental model"
In principle I agree but the difficult is getting to the user's mental model. Often, people simply cannot describe it accurately themselves. It needs to be teased out with long user interviews and testing, ideally. This can be difficult to arrange.
|
|
|
|
|
Because while you can get a degree in Computer Science, all the way up to Piled Higher and Deeper, very few people actually are skilled at the art of software development.
|
|
|
|
|
Because programmers rely on the latest and greatest tools as marketed by salesmen who, in their previous incarnation on earth, were selling snake oil.
|
|
|
|
|
This means that installing WSL is now easier than ever, as now when you run wsl --install all the necessary components that you need for WSL will be automatically enabled, including your specific Linux distro of choice. Making it even easier for this to be The Year of Linux (on WIndows)
|
|
|
|
|
Kent Sharkey wrote: install in Windows 10 insiders preview build 20246 They can't manage to make the release version stable or without messing up with something... no way I'm getting close to the insider preview build
M.D.V.
If something has a solution... Why do we have to worry about?. If it has no solution... For what reason do we have to worry about?
Help me to understand what I'm saying, and I'll explain it better to you
Rating helpful answers is nice, but saying thanks can be even nicer.
|
|
|
|
|
Starting from November 5, Windows 10 users will see a clear distinction between automatic and manual updates in Windows Update. Microsoft says this change will allow users to have more control over plug-and-play accessories that automatically download and install drivers. Was "Plug and Play" every anything other than "issues"?
|
|
|
|
|
Kent Sharkey wrote: a clear distinction between automatic and manual updates in Windows Update. Are the manual ones, those that allow you to choose "restart later"?
M.D.V.
If something has a solution... Why do we have to worry about?. If it has no solution... For what reason do we have to worry about?
Help me to understand what I'm saying, and I'll explain it better to you
Rating helpful answers is nice, but saying thanks can be even nicer.
|
|
|
|
|
Security flaw lets attackers escape sandboxes designed to contain malicious code. Releasing information about it certainly will stop people from using it
oy
|
|
|
|
|
Kent Sharkey wrote: Releasing information about it certainly will stop people from using it and not to forget to limit it to very few people...
Kent Sharkey wrote: Security flaw lets attackers escape sandboxes designed to contain malicious code. That's why I think I'll keep using VirtualBox or VMWare
M.D.V.
If something has a solution... Why do we have to worry about?. If it has no solution... For what reason do we have to worry about?
Help me to understand what I'm saying, and I'll explain it better to you
Rating helpful answers is nice, but saying thanks can be even nicer.
|
|
|
|
|
|
The object recognition technology could clearly do with a bit of a tweak, or else the team might actually have to implement the policy of bald referees being forced to wear a sombrero to differentiate themselves from the ball.
|
|
|
|
|
The fix will work until someone moons the camera.
|
|
|
|
|
AI Shmay I, My dog can do better than that.
|
|
|
|
|
Just imagine they'd use that AI in football playing robots...
Fortunately, I do not play football at all.
Oh sanctissimi Wilhelmus, Theodorus, et Fredericus!
|
|
|
|
|
Google says hardware in embedded devices needs to improve to make possible a world of peel-and-stick sensors free of wall power and human maintenance. It's all fun and games until you have to apply a patch to trillions of devices
|
|
|
|
|
We need a name for these. How about Replicators? (Yes, I'm binge watching Stargate SG-1)
|
|
|
|
|