While you are likely trolling, no one is suggesting failing companies should be propped up with public money. But successful companies still often have many failed projects, and this is especially true for any company which relies heavily on R&D. Like they say, if you aren't failing you aren't innovating. A solvent company needs to pay for its failed R&D projects with its successful projects or it would go under.

It is very similar to running a VC company. They obviously don't think ever venture will be profitable, but they are hoping to make an overall profit on those 20% of companies that actually succeed.

I assume it takes a month because all of the background checks take two weeks and then they have to wait for the employee to give the current employer a two week notice. That is what happened in a company change I made earlier this year, although it only took a week for the background checks.

If you look at the listing of 20 tech positions, the Software Engineer position is a strange outlier. Its 35 day duration is almost 7 days higher than the #2 position, Senior Applications Developer, which is 28.3 days. The rest of the time-to-hire durations group together much nicer, which the overall trend being more senior positions taking longer and entry level positions taking less time.

So the duration for senior level tech positions appears to be around 27-28 days, which is what the summary should have focused on.

I originally thought the warning meant pregnant women are dangerous.

Yes.

Sorry I guess I am a bit too old and still don't think of mobile when I say cross-platform. But not old enough to immediately think embedded when saying cross-platform. I have been living purely in the desktop / server world for over a decade and that is where my project resides as well.

The project is very similar to writing an database management system. I didn't want to get too much into the details so people aren't commenting about the virtues of the project itself.

There is no UI component; it could be thought of as a cloud service.

My perfect solution would be developing it in C# while having complete control over memory allocation and release. I have done extensive testing using System.GC.Collect() to manually control garbage collection with no luck.

I started programming with C++ in high school and began my career with the language so I'm not too worried about moving back to the language, but I realize I am far more competent with higher level languages right now.

The only thing stopping me from doing this now is I need the ability to release memory on demand, and System.GC.Collect() is not sufficient. I really wish it was since I would work in C# if it was sufficient, but all prototyping I have done so far has not been successful enough.

I started my career working with C++ so I am not new to the language, but my work used very little of even the standard library let alone other third party tools. I used in house libraries that were already written. I was also a novice at the time, so a good deal of the last few months has been spent reading material like Effective Modern C++ and others in the Effective C++ series, along with a throwaway project to practice while reading.

I am a firm believer that most performance is in the algorithms not the language, so I am definitely not choosing C++ primarily for performance reasons. I need certain aspects like memory management, but my choice is not because I think my algorithms will run faster in C++ vs Java. There will probably be some very important modules I need to optimize, but all of that will likely be long after I have a working prototype.

I had a project a year ago where tokenizing terms needed to be as fast as possible, so I tried creating a C library to do it. I am confident I did it well by not copying memory but just keeping track of the location and length of each word in the existing character array, but I still couldn't match the performance of just doing it simply in LINQ (although it was almost identical performance). There is likely some kind of CPU caching or other tricks I was not doing right, but my point is I understand moving to C++ is not some silver bullet for performance, which is what I meant when I said I wasn't moving to C++ for performance reasons.

I can and have read up on Boost::Asio and a little on QT, which is why I know how they would be useful to me in this project. But I am looking for advice from people who have used them extensively and know the gotchas. In my experience it takes a couple years to really know how certain aspects of a framework will affect your project long term and I want to gleam as much insight from the community as possible before I start down a certain path. It may not be possible to learn anything substantial with one Ask Slashdot post, but I figured why not try.

Okay, basically I need to be able to allocate and release memory manually and without waiting for any garbage collection. I need complete control over concurrency and memory sharing, and as little overhead as possible when accessing the hard disk. I have not been able to find a way during prototyping to control memory enough in Java or C#. A language with a great cross platform library but no memory management would be perfect, and right now C++ is the only language I know of that comes close to those requirements.

The project is very similar to writing an database management system. I didn't want to get too much into the details so people aren't commenting about the virtues of the project itself.

By granular level I mean I cannot be in a managed environment like the JVM or .Net JIT compiler. I need to be able to allocate and release memory manually. I have done some prototyping in Java and C# hoping I could control garbage collection enough for my needs, but it isn't possible (or at least I can't figure it out).

There is no UI and it will only communicate via networking. It will also be distributed and concurrency control will be important.

There is no UI required for the project, although I realize you can use modules like QtNetwork without the UI libraries. I am a bit worried about tying myself too closely to the QT event loop. Can anyone provide any insight on if the event loop will complicate concurrency throughout the rest of the application or if it is even necessary? I have briefly read through some documentation but I am most looking for advice from someone who has used it before and has dealt with the gotchas that every framework has.

I have leaned more towards trying Boost over QT mostly because it relates closer to the standard library and has less overhead and framework buy in necessary. Or at least that is my impression. Has anyone tried both and have any advice on which they prefer for a project where you don't want to completely drink the framework's Kool-Aid?

You are either very young, in a low paying field, not very good at your job, or live in rural Kentucky if you think $600 per day is extortion. That is not even $100 per hour. I make that as a software developer as an FTE even without including benefits in the midwest in my mid-30's. And my employer is getting a great deal and knows it, which is why I am paid extra for overtime which includes either extra bonus or extra PTO days to compensate. My young daughter and second on the way are what keeps me from taking more risks in my career and probably making much more money (emphasis on "probably").

Come to think of it I am probably just wasting my time here because you are likely just trolling yourself.

I actually list C/C++ on my CV purely to imply I am not strong in either. It was my main language in college and I have used it sparingly in my career, but I don't consider myself a C or C++ developer. So if you are looking for a senior C or C++ developer, look somewhere else. But if you want someone who at least understands pointer arithmetic and/or may have to lead a project where some of the developers are working in C or C++, then I may be your guy. I always clarify if asked, and never apply for jobs where I know they need a C++ guru anyway.

Do you feel the same way about Boost::Asio as you do about QT? I am willing to do more low-level coding when necessary; I just want to spend as much time as possible on core functionality. Network code does scare me a bit since I don't remember doing socket programming fondly in the 90's, but I'm sure I could figure it out. I do love the idea of a library doing it for me, although don't want the library to spread too far into my code. I want to have strong control over concurrency and memory management and don't want the library to limit me in either area.

Thanks for your input.

I had thought of QT as a GUI framework, but after a few people here have mentioned it I see it is another contender for a cross-platform C++ library to match my needs. Thanks for the info.

Do you have any opinion on the benefits of QtNetwork vs Boost::Asio?

Obviously your altruism doesn't have to change when you have kids, but a family budget without kids is grossly different than one with kids. All of a sudden you are paying $1700 per month for day care, or you are living purely on one income. Because you are the sole breadwinner, you need $150 per month for life insurance (if you are healthy; much more if you aren't).

Then you realize the home you live in isn't in that great of a school district, and you find there are no cheap homes in good school districts (done on purpose with zoning so property taxes per child are high). Then your housing goes from $1500 per month to $2500 per month. Then you realize college already costs $25k per year at state schools now, so you better save $400 per month for your kid's college education. The cost of clothes and toys and child furniture starts to grow, and you realize how expensive supplemental formula and diapers are.

It adds up.

I obviously cannot claim we couldn't donate over 10% of our income if we tried. We shop at Brooks Brothers and Lululemon, have one nice $200-300 dinner-date per month, and buy the latest phone & iPad every other year. We rationalize it as being necessary to stand out as elite professionals to keep our incomes growing, which is party true but mostly self-serving. We buy mostly organic foods, and an extreme amount of fruit to satisfy my sweet tooth without buying Skittles (organic grapes are not cheap).

But then again we spend almost $30,000 per year in federal income taxes after all our deductions, which is about $20,000 more than the average US household. So I guess we do give about 10% of our income in donations; its just that the government would force us to even if we didn't want to.