|
Ask them to sit with developers and make the damn notes. Real time!
If you can't trust your team, you should not be working on whatever it is in first place. Yes, status checks and monitoring are important but these minute details, no.
"You'd have to be a floating database guru clad in a white toga and ghandi level of sereneness to fix this goddamn clusterfuck.", BruceN[ ^]
|
|
|
|
|
The tasks should be broken down in something like Jira. Given appropriate states, it becomes trivial to see what the status of tasks are without the need for meetings. The key is to keep tasks small.
This space for rent
|
|
|
|
|
Yeah we are breaking down the tasks as much as possible. This particular thing comes into picture, when the nature of task is like Writing algorithm for a module.
For continuous days, the guy in context, will not touch anything, but on third day, he completes it matching with the estimate that he initially given :3 days. And it's something worth 3 days in fact. The first two days, he got nothing to share in the Sync meetings. He just says, I'm studying the problem, That's WIP, with a sheepish looks. But he goes back to his desk & spend hell lot of time in Facebook. I think somewhere in Facebook he's taking inspiration from to get into his work by mid-night of 2nd day, before coming to office.
Why I'm not much affected by this usually is, The type-1 people who rigorously work, for sure take 3 days, to finish the same problem. The only thing I suffer a bit is the anxiety he creates by the 3rd day, I always doubted if he'll be able to finish it on time. 95% he's finishing it by 'EOD'. As he calls it! 'EOD' means precisely 12:00 am in his language. Not the evening 7:00, that all people log out of work.
Starting to think people post kid pics in their profiles because that was the last time they were cute - Jeremy.
|
|
|
|
|
I believe that while it is difficult to tap into the subconscious, the subconscious is aware of problems we are contemplating, works on them, and can "surface" a solution to our conscious state as inspiration. So, yeah, "ignoring" the problem for a time actually means it's being worked on, but not consciously.
|
|
|
|
|
A bit of caution about Pete's approach:
If you take that approach, you might get dangerously close to an environment where you can trust the development staff to manage themselves.
Then, the organization might realize that management is overrated. Who will we get to synergize efforts across problem spaces buzzword buzzword buzzword?
|
|
|
|
|
Your directors are wrong.
|
|
|
|
|
I had a boss who wanted a daily standup where we reported status. We held it on an IRC channel instead of face-to-face (we had a profoundly deaf colleague). I got in the habit of pre-writing my status report and pasting it into the IRC window at the exact start of the meeting. My colleagues began to do the same. We achieved a 60-second standup. Once or twice a month, we actually had something to talk about, like needing help or hitting a roadblock. We managed to convert a useless meeting to a meeting whose uselessness took so little time that it evolved to a purposeful meeting.
I had another job where we had a daily standup where people reported status. I started emailing the group my status before the meeting. This caught on among the group in about one week. In this way we took a "standup" that took one hour and created a more reasonable 15 minute meeting.
Never do in a live meeting anything that can be done with electronic communication. The cost is too high.
And yeah, you can call yourself agile, and describe your interminable meetings as standups, but if you do, you are delusional. Most managers are delusional. Why is that?
|
|
|
|
|
Only ask about problems / impediments "during" the week; report "progress" at the "end of the week".
Most problems are caused by not enough partitioning (i.e. tasks requiring more than 4 man-days); but expecting a "deliverable" EVERY DAY is ridiculous.
I always gave people a week to complete a task or two; and the next week to fix it in case there was an actual problem and get back on schedule. Never failed. And involving folks in the estimating process usually drives the whole process to a successful conclusion because it was "their" estimate (no matter how ambitious it might have been).
There comes a time when one deals with the peanut gallery and takes ownership of the project (if one has the confidence and cares enough).
|
|
|
|
|
For the last ten years newcomers choose Java/C# over c++ is mainly because they both come with comprehensive libraries that can fulfill most of their requirements.
In C++ we have std and boost, but in some point they are confusing and their functionalities are somehow overlapped, which introduced a much harder learning experience compared to other languages.
The closest framework I know is Qt, but then Qt isn't open source.
|
|
|
|
|
.jpg wrote: For the last ten years newcomers choose Java/C# over c++ is mainly because they both come with comprehensive libraries
Mainly because there are courses offered for the first, and not so many for the latter. There's also a difference in available positions for both.
Bastard Programmer from Hell 
If you can't read my code, try converting it here[^]
|
|
|
|
|
They once made a solid , comprehensive library after carefully studying all the pitfalls in C++,
and they named it .net.
To be honest, C++'s role is narrowed down today towards System's application. Even there's it's got competition from Java, CE.net etc, & even Arduino is making a threat. There were so many biz applications running on C++, but one by one, everything is moved to .net or Java. I got quite a number of friends who were working on all these migration projects in Banks, Insurance companies et all.
Even , one of the complex projects I worked on, we split the project into two and let only the core run on C++ , everything else runs on Managed C++.
So what is your intention for learning C++?
Starting to think people post kid pics in their profiles because that was the last time they were cute - Jeremy.
|
|
|
|
|
.jpg wrote: In C++ we have std and boost
We, usually, also have access to the full platform API.
.jpg wrote: but in some point they are confusing and their functionalities are somehow overlapped
It's not unheard of for something to start out as boost library; and then, at a later point in time, serving as the foundation for what makes it into the C++ standard library.
Espen Harlinn
Chief Architect - Powel AS
Projects promoting programming in "natural language" are intrinsically doomed to fail. Edsger W.Dijkstra
|
|
|
|
|
Yep...examples include boost::filesystem (a version of which will become std::filesystem in C++ 17), or boost::thread (which developed into std::thread in C++ 11).
Anna ( @annajayne)
Tech Blog | Visual Lint
"Why would anyone prefer to wield a weapon that takes both hands at once, when they could use a lighter (and obviously superior) weapon that allows you to wield multiple ones at a time, and thus supports multi-paradigm carnage?"
|
|
|
|
|
Quote: It's not unheard of for something to start out as boost library; and then, at a later point in time, serving as the foundation for what makes it into the C++ standard library.
That's because the original purpose of Boost was as a "proving ground" for features to be added to the Standard library.
Truth,
James
|
|
|
|
|
.jpg wrote: The closest framework I know is Qt, but then Qt isn't open source.
Since when? Qt - Download Open Source[^]
Jeremy Falcon
|
|
|
|
|
 lol
"On 20 May 1995 Troll Tech publicly release Qt 0.90 for X11/Linux with the source code under the FreeQt license."
Starting to think people post kid pics in their profiles because that was the last time they were cute - Jeremy.
|
|
|
|
|
|
It does rather depend what type of functionality you are looking for.
Aside from the usual suspects (Qt, WxWidgets etc.) there is no single "framework", but there are some amazing cross platform libraries. For example, in addition to the Standard library and Boost, I can recommend POCO[^], STLSoft[^] and Xerces[^].
The biggest challenge is building and deploying them as there's no unifying build/deployment mechanism for C++ libraries (we use a bunch of custom scripts). However, there are a few package managers (e.g. Conda[^], Conan[^] or Pacm[^]) which are trying to make this easier.
Anna ( @annajayne)
Tech Blog | Visual Lint
"Why would anyone prefer to wield a weapon that takes both hands at once, when they could use a lighter (and obviously superior) weapon that allows you to wield multiple ones at a time, and thus supports multi-paradigm carnage?"
|
|
|
|
|
What's wrong with just using C++/CLI to use .Net?
|
|
|
|
|
I think looking for a single comprehensive library for C++ misses the point of C++. In fact, that is what spawned Java in the first place which captured popularity by offering automatic garbage collection which was not baked into C++. C# was merely a "me too" programming language to take on Java. But as Bjarne Stroustrup says in one of his books, there are a dozen such garbage collection libraries to be had in C++. The same goes for GUI libraries (eg, I use BCGSoft's GUI lib and not Qt). I also have an encryption lib I subscribe to. And for other parts of the code, I write my own managers that tend to look like library pre-cursors. But the main reason I use C++ exclusively is, I'm not beholden to a single corporation's fickleness. C++ is internationally standardized which means there are 100 Computer Science professors at any one time making dead sure the language is relevant, fast and correct. Each chosen library vendor I work with is chosen out of deep respect for their company's longevity and integrity of their product. The pennies I pay for their libs would have cost me dollars to write my self assuming I could.
modified 25-Mar-16 8:27am.
|
|
|
|
|
Yeah. What he said. A cross-platform library (for instance Java Swing) is always limited because it becomes the least common denominator of all the platforms it covers. A native library (like .NET) is locked to a single platform, often by design.
The success of the C++ standard library results from the very limited extent to which it depends on an underlying operating system. Most parts only require a memory allocator. A very comprehensive library would require much more support, making C++ less viable on embedded platforms. As it is now, you have a choice of several libraries, depending as you are willing to accept cross platform limitations or embrace a single target.
Win32 is a real solid and comprehensive library for C++. There, problem solved. Oh wait, I bet you wanted free and open source too. Yeah, there's another reason why there is no library. Who's going to spend millions to write it for you?
|
|
|
|
|
All the "comprehensive" libraries in other languages derive from c/c++ if not being just a wrapper around it. They do not offer anything beyond what c/c++ offers and sometimes offer less. The appeal of those languages stems from IDE that takes care of project management and often offer direct access to UI functionality. The problem arises if one needs platform/GPU computing access, anything more complex requiring native code or just plain control over code execution and memory management. In those instances native c/c++ shines.
|
|
|
|
|
I've grown quite fond of JUCE which is pretty much a library that covers most aspects you'd need. And cross platform, of course.
|
|
|
|
|
⎺⎻⎼⎽⎻⎴⎴⏜⏞⏠ (10)
Not that you actually need it but I'll confirm that the answer is DEPRESSION, the clue being an illustration of a depression in a surface followed by a number of downturned mouths.
Same bat-time, same bat-channel, tomorrow then. 
I am not a number. I am a ... no, wait!
modified 24-Mar-16 11:34am.
|
|
|
|
|
Is it a normal British road surface these days?
Bad command or file name. Bad, bad command! Sit! Stay! Staaaay...
|
|
|
|
General 
News 
Suggestion 
Question 
Bug 
Answer 
Joke 
Praise 
Rant 
Admin 
Submit an article or tip
