So: I hope that's the optimistic sign it might be ;-)
2,199 karma · joined September 29, 2009
I wrote Rebuilding Rails, which I think is pretty good: http://rebuilding-rails.com
Some stuff I've done: https://codefol.io https://github.com/noahgibbs https://codefol.io/portfolio
I'm not hard to Google for, if you want to.
So: I hope that's the optimistic sign it might be ;-)
I've been a software engineer for going on 20 years now. Have a plan, just in case, even if you're good at programming (I am.)
You didn't used to have to exclude the server market. They owned that too. Then Linux challenged them and won.
In fact, they used to own all PCs. Then Apple challenged them for laptops and won.
Microsoft isn't doing badly, but it's not an unchallengeable juggernaut any more.
And Google won't be forever either.
There are other games in town, but those two games were seriously inflating salaries here.
That would be bad if the content changed, but in some cases you can be sure it won't.
With that said, I'm not 100% sure what it's doing for "normal" CDN files either. What you're saying sounds like a flaw, and I don't know if I don't see the obvious answer, or if you're right and that's a significant problem.
They could stay a major player for a long time and do just fine, but they'd go from being "Apple under Jobs" to "Microsoft post-Gates." It's a long, slow sunset, and you sleep fine at night, but you don't win it all.
With that said: yup, the Apple Watch does put off the final sunset a few years.
You can go manager or freelancer, of course.
We could do something that would be considered an act of war in order to kidnap somebody explicitly protected by the target country. But these things have been known to go poorly.
Neurological surgeons had the highest median daily rate for providing on-call coverage, about $2,000 a day. Near the top of the pay scale were neurologists ($1,500), cardiovascular surgeons ($1,600), internists ($1,050), and anesthesiologists ($800).
Among the specialists earning lower median daily rates for on-call compensation were: psychiatry ($500), general surgery ($500) gastroenterology ($500), ophthalmology ($300), and family medicine without obstetrics ($300), according to the MGMA survey data.
But in general, "salaried" has not been construed to mean "the company owns every minute outside the office."
Folks like Fred Wilson and PG are doing their part to give the impression that everything is open and kinda-sorta known, which wasn't true 15 years ago.
The fact that it's still about equally sleazy then comes as kind of a shock.
In other words, if it's genuinely side work.
There's a specific statute to that effect, and they're required to inform you of same.
(I'm not a lawyer either, but I've worked in CA for a long time.)
It's explicitly called out in a lot of business books -- as a small business you often have a lot of cover to undercut the incumbent, because they're just not interested in a niche that may be huge from your point of view.
Of course, this is at least a terrible summary article, not clear if it's a good study.
The equivalent for jobs wouldn't be better.
GlassDoor makes it easy for that reputation to spread.
Companies basically never do that, I assume because if you then don't do anything about the tension, employees get the attitude, "oh, management already knows about the problems and doesn't care. Guess I won't worry about doing better."
Of course, the proposed solution here is to just make the tensions public, which seems like it would fall instantly into that trap.
I could say "the main difference between Ruby 1.8 and 1.9 is that programs run twice as fast in 1.9" (true, 1.8 is dog-slow.) And that's "Y is better", but it's also fairly specific.
"Python 3.0 makes it easier to produce high-quality software" is annoyingly vague. They seem to mean software that behaves more consistently by reducing the amount of default unpredictability in the language spec -- that's fair and specific. It's also not what the article said, sadly.
His example is using it as an angle, where comparing two of them is conceptually pretty sensible.
If you pretend it's a structure, sure, you'll need to extract primitive data types out of it before doing anything with it. But if you want it to behave as a higher-level abstraction, reimplementing less-than (say, for comparing the vector magnitude) at every call site isn't great.