Plus from a pure software perspective, users usually hate change, even if it's good. I can't imagine sneakily enforcing change on several million people could at all go over well unless you have a REALLY amazing product and from what I hear 10 is just pretty okay.

So I know absolutely nothing about Linux, how does Mint relate to Ubuntu? Are they forked from some base code years ago? Not related at all aside from linux kernal?

I am having a lot of problems with programs crashing in Ubuntu so I was looking for a different Distro, but am I just going to have similar problems on other ones I guess is the question?

The fewer employees they have working legacy support for 7, the more they can have making 10 better. Makes sense they'd try to push people to 10.

Also, if average users just think it's 'pretty okay', then you can assume there's a lot of features they don't know about or understand that makes it a lot better. SSD handling got lots of perks. Is that apparent to the average user? Of course not.

Mint has two versions. One based on Debian and one based on Ubuntu. Ubuntu is also based on Debian. Ubuntu / Canonical also gave your data to Amazon or something like that years ago. Being privacy-conscience, I will never trust Ubuntu again.

In any case, mint isn't a "desktop," it is a linux version. The default desktop is cinnamon, I think, and you can install cinnamon on any distro you want.

Really, the big difference between distros are the package managers. I personally think apt-get is a horrible package manager. I'd much prefer using dnf or pacman.

The second difference is the distro's ideals about being totally FLOSS. Mint is a little more lenient on this ideas. Fedora is about as hard-core anti-propriety as you will find among the popular distros. With that said, I have Flash on Fedora (isn't hard to do). I haven't put Chromium on it (yes, they are that hardcore). I know I've trumpeted it before, but Fedora is a solid OS. Never had an issue with programs crashing.

But when Microsoft kept supporting old versions of their software, developers were constantly complaining about having to support IE 5/6/7/8/9/10 and how Microsoft should force people off these old browsers and so on. And for that matter, we're getting these complaints about Android and how it makes our lives hard that there are so many different versions of Android in the wild. They can't win.

Yeah supporting legacy products at Microsoft is very expensive. That company is mega rich though, awesome really. However, I pretty much agree with JMakin though.

They have always been about legacy systems, arguably to a fault. This move is really aggressive.

OS X Mountain Lion, which is newer than Windows 7 and closer to Windows 8 in terms of release date, is already unsupported. ChromeOS has always updated automatically.

What they are doing here is switching the consumer segment to an auto-update schedule because they can't be expected to maintain a legacy system free of malware by themselves, make Microsoft look bad by being tied to old products and virtually never pay for software anyway, while charging a premium for the enterprise segment because they are willing to pay much more to make sure that everything is still supported.

nm

If I'm on W8, is it a no brainer to upgrade to 10? I play games and heard occasional stories about incompatibility issues/slow alt-tabbing so I've stuck with the status quo (where neither of those things are problems) for now.

Pretty much. Windows 10 is probably going to be better supported than Windows 8 going forward by third-party developers anyway, given the adoption curve:

http://www.cnet.com/news/windows-10-...top-os-market/

Staying in Windows 7 has its logic but sticking with Windows 8/8.1 doesn't really make a lot of sense outside of specific compatibility scenarios.

I have a 6-ish year-old poker desktop PC I'm trying to decide if it's worth it to upgrade to W10.

I think I had to use 7 for a little while but then I went all Mac and have been much happier...

I am on 8.1 (imo best windows once you install classicshell), and really do not want 10 forced upon me. I don't want even 1 day of downtime, while I am happy with what I've currently got.

Not sure what's bad here, my code or socket.io:

usCountyData is a ~31mb string (I know, I know, but it's more fun to actually do cool **** with the data first and worry about how to send it efficiently later), and shortly after it receives a map data request, my Node.js instance blows up to about 1.5GB of memory usage and crashes.

hmm I would equate wondering why your websocket is having trouble with a 31mb transfer with wondering why you're having trouble driving around your 100ft long tractor trailer. Technically you can do it but...

i'd post on SO, but if I had to take a guess websockets is freaking out somehow because of the file size. since usCountyData (i assume) is a static file, i'd try sending that part over ordinary http, which makes more sense anyway.

I think I may be looking for a new job or otherwise pursue a new opportunity over the next year or so. It's funny I feel like I know in the abstract exactly what I need to do but it's been 9 years since I last interviewed for a job so it's a weird feeling. The timing seems less than ideal given that I think we're at or just past the cyclical peak but I just feel like my life is passing by.

Do practice interviews at places you don't want hired at or with friends imo

d3.js is proving to be a really good library to create data visualizations on the web.

We're doing something pretty cool with it right now and going to be doing several more things with it in the coming weeks.

If you are looking to create any type of visual display for data, it has a lot to offer.

My friend and I were discussing this yday actually, he was saying it looked like it would take forever to learn to be functional using it. Is that at all the case?

Thanks! I think I do better when I wing it but either way I have to figure out what I want to do first and also how to transition away from the current situation and make sure everything is in good hands (that alone will take 3-4 months) and so on. Currently, I have no idea if I want something technical, management-related or even finance. Or if I should start my own business and if so, exactly what.

Our strategy is to shorten the learning curve by finding sample projects that we can start with and build on top of.

Starting from scratch seems daunting, but if you look through the examples you are very likely imo to find something useful.

https://github.com/mbostock/d3/wiki/Gallery

I'm actually pulling from a postgres database. There are some things I can do - compress with zlib before sending, and I'm looking into how to tile-ize the data so it's less detailed when zoomed farther out, but it's less fun to have to do the optimizations like that up front.

New broom at MS. With their cloud services (which seem pretty good), OS strategy and move to expand development tools to Android and iOS, they are significantly changing direction.

While it's a strategy that is going to upset some, I can see how it makes sense for them.