Yeah, but it's dead, just stuck on repeat nailed to its perch.
As is the old "user interface hall of shame" - dead but archived in various places.

And yet every damned thing they said about UI design is still correct and relevant - it's just that no one actually gives a s**t anymore. Success in UI is now measured by how long you can keep users on your site / in your app, not how fast or how accurately users can actually get stuff done. Or how many ads you can show per user action.

Eventually the decline in UI will render the user irrelevant - when computers start doing things for the user it won't be because computers are smarter, it'll be because even a dumb computer can do a better job than the user through-the-crap-interface. Then we can all climb back into the trees and watch the computers screw up the planet.

Fundamental problem is that to do this best you need a product that is either:
(a) too complicated or too badly documented to install/configure/use without help
or
(b) too broken for anyone to dare use it without a support contract and SLA

Example:
Q. I've wasted hours trying to get the community edition working because of lousy docs, anyone want to work together on producing some better ones?
A. This would be a waste of (even more) time, because the paid edition already has excellent docs, don't waste your time, just pay

The moral problem that is still here is that your revenue stream depends on, often knowingly and deliberately, producing crapware (whether it's the docs or product or whatever). In contrast, producing excellent well-documented software that is easy for anyone to install configure and use gets you zero revenue.

You're right, BUT, I think it is your career that has evolved, you've become a consultant. Not a bad thing, bill the hours and any code (or not - end product may be advice to buy X or configure Y to do Z) is a side effect. You may also be able to release code/solutions for free, unsupported, off the back of paid work - many customers don't care as long as they get their solution and (non-exclusive) rights to any custom code in it. Exclusive rights cost more - because you can't use the knowledge gained or the code for another customer - most people don't want to pay for it.

However, you tend to get to this point by gaining experience in various jobs and building a reputation over years. The submitter sounds like someone starting out, probably without much of a clue. You can't go straight to freelance consultant at E70/hr from there.

The real answer is (a) get a job (most software dev/IT is still custom/internal code not sold proprietary code) and (b) do FOSS dev as a hobby. When you get enough reputation from (a) or (b) (preferably both), you can move on to your become-a-freelance-consultant solution.

Need to distinguish carrier-lockin from manufacturer lock-in - they are different.

If it is a phone with a "Carriers adapted custom versions of Android" then you don't actually own the _device_ either - you are renting it until it is paid off on contract and many people never even do that because they get a carrier-provided "upgrade" and carry on renting. If you _buy_ your phones sim-free _then_ you _own_ the device, and guess what - no carrier-special garbage on it, no carrier lock-in, no carrier-special app stores.

Manufacturer lock-in is a different problem, still a problem but there are multiple mfrs to choose from and at least buying sim-free you can be reasonably sure you are getting the same phone as in the reviews (where the mfr has some incentive to show good performance) rather than a carrier-crippled mess.

The problem with carrier lock-in is that most people are too dumb, too disorganised with money (note, not too poor - renting costs more) or simply too seduced by shiny-stuff they can't really afford (upgrades...) to escape the carrier phone rental model. The same people are also typically tied into phone contracts, paying way too much for their mobile minutes/data and unable to switch carriers or change contract to suit life changes - this is not a coincidence...

Me, I am happily running the family phones on sim-only 30-day contracts at less than £10 a month, most of the phones are over 3yrs old and still doing as good as they did when new and doing what the users need.

In fact the "other countries" have already made laws, in this case the EU has privacy and data protection laws which mean MS cannot hand over the data without being in breach.

What is seemingly obviously required is international agreement so that the US can request the data from the relevant local jurisdiction who will be able to get it under the relevant local laws. Funnily enough, such an agreements (mutual legal assistance treaty) also already exist, and the EU laws which prevent MS handing it over also allow for access by local law enforcement subject to local laws.

That is the real story, the US wants to get the data by the back door because it can't be arsed to do the job properly and use MLAT to (effectively) get a court order in the EU / Ireland. You can take your pick as to _why_ they want to do it this way from e.g.:

a) the US is too lazy or incompetent to seek a foreign court order
b) the US doesn't have evidence sufficient to convince a foreign court that an order would be required (add in your own take on how it managed to convince a US court...)
c) the US believes on principle that it has jurisdiction everywhere so it shouldn't have to go to a foreign court, despite having signed treaties to facilitate exactly that

Um, Google pulled (at least pulled google news) out of Spain (which is both first world and in-EU last time I looked) for precisely this reason.

Arguing they won't do something they've already done before for the same reason, makes no sense whatsoever.

Um, you know MS had exactly this sort of technology on Windows back in the early 90s right? When Windows NT ran on DEC Alphas as well?

MS arguably did it better - they didn't need to emulate the OS because they could recompile it for Alpha (as they do now for ARM). Win NT has always been a portable OS. Apple only had to emulate OS level stuff because they wrote hardware-specific stuff into the OS in the first place.

fx!32 didn't just dynamically translate, it also cached bits of native code translations for subsequent runs. We had some alphas on demo at the time and they were beautiful machines, didn't notice much, if any, performance hit from fx!32 - in fact I think we stopped bothering to recompile our applications and just ran the x86 versions because there was no benefit in it.

The real problem was that Alpha (sadly) failed as a platform. The issue has never been that MS doesn't have a portable OS, the issue has always been that they are on by far the most popular hardware platform already, releasing and supporting a port costs (big) money and for what? The landscape is now changed, Intel have ditched their low power ARM competitor (Atom), so ARM is now king in low power with no other contenders, there is incentive to port.

In fact if you look at the previous attempt (windows RT) and compare with Intel releases, it looks a lot like Windows RT's purpose was to kick Intel into delivering better performance on Atom (they did, and RT promptly got shelved).

There is no connection _if_ you use separate mail and calendar services, and you don't connect them.

There is no connection required unless you want to do things like, oh, send meeting invites or update calendars based on invite responses... say. Obviously if you only ever meet yourself, that's not a problem.

If you want to do that, then you need to connect your mail and calendar services with, say, an iMIP service/connector - you know, the standard specifically for connecting (non-MS) mail and calendar services. Exchange just provides a mail and calendar service that is pre-connected without installing and configuring a connector, Outlook provides a client that shows both. Interestingly if you don't use Outlook, the built-in Windows email and calendar apps are separate - or rather they appear so, but you can't send meeting invites from Calendar until you configure an account in Mail...

> There is the third possibility in this case: it is homeopathy, intended to be diluted, with a manufacturing defect.

If there is any active ingredient in it, it is not homeopathy, it is drugs. The FDA can then shut them down as a drugs manufacturer making stuff that has not gone through testing and approval. Homeopathy doesn't need testing and approval because there is no active ingredient in it, drugs do, because there is.

Manufacturing defect? - Homeopathy's manufacturing process is defective by design, the entire point is to remove anything that could possibly have any effect.

Can't have it both ways - if the stuff is diluted so much that there is nothing there, how the f**k can it be dangerous and need to be recalled?
The reason the FDA can't shut homeopathy down is they are selling nothing but sugar pills, any effects or side effects would be the same with a placebo.

The _only_ danger in homeopathy is people stupidly not getting proper medical treatment because they are using it - will you can't regulate away every form of stupid. Note that you can, and they do, shutdown any practitioners who advise patients to not have effective medical treatment and to use snake oil instead.

There are two possibilities in this case:
1. it's _not_ homeopathy and actually has some active ingredient in it, and it's not a nice one
2. it is homeopathy and therefore it doesn't have any active ingredient in it and therefore it is _not_ dangerous (or effective)

It _cannot_ be both homeopathic and dangerous.

In the UK most employees don't file tax returns, your new employer gets (part of) the P45 form from your old employer, in order to work out your tax coding for the rest of the tax year - the information on there is enough to work out your previous salary.

Now, you don't _have_ to do this way (can use P46 I think instead) - but then your tax will be wrong, which can get expensive.

If you think the cap is high, just look at what they charge overseas students (and yes, STEM courses are much higher)

http://www.undergraduate.study...

True, you can get all the research papers you need online these days, just log in with your uni id to get access. If you've ever wondered why it costs $$$ to get the same papers when you're out of university, well the STEM journal publishers give all university students free access, obviously...

Yes on all counts. At the same time.

I ran the dev team, already had experience of very early agile predecessors (DSDM), I was told to "do" agile with my team and went into it as a confirmed sceptic trying hard to keep an open mind, and f*** me it worked. Well. Really well. Several people (inc me) thought we couldn't do product development agile, couldn't do support and maintenance agile, couldn't support pre-sales agile, well we did, and it worked better than anything else we did in a whole lot of ways.

I have also since seen the same agile process (SCRUM) used so badly it drowned a project in treacle, it would have been hilarious if it wasn't for the amount of money you were watching being pissed up against a wall.

Agile is a tool, it is a better lathe, used properly it can produce things of great beauty and precision, but it is still just a better lathe - if you are trying to use it to drill a hole or cut a straight line you will still fail, if you leave the chuck key in you will still get hit in the nuts (if you're lucky) when you start it. Know your tools.