csmithuk

It's unfortunately also very CPU intensive which is a big problem for IRC networks which handle stupid numbers of connections at the same time.

Yes. Lego, computers, electronic bits and kits to make, chemistry set, things to take apart, piles of technical books plus my father was a major clone PC importer and manufacturer for over a decade and our house was literally up to the ceiling with PC parts and software.

However, with none of that I think I'd have done the same thing.

It's not all that bad now and wasn't when I was at school (in the UK). During my time, we had the Acorn BBC Micro to start with, then the Archimedes. These were both programmers dream machines and code we did, usually plugged into large Lego machines. After that it diverged into "office studies" as RM dumped millions of PC clones but there was still Quick Basic and Turbo Pascal available that was taught at a lot of schools either as a mainstream subject or through "computer clubs".

Now there was a gap for me but I have children now and they are learning how to write HTML and basic JavaScript. They are using software to produce video productions and stop-frame animations, they are even getting instruction from parents on how to write python. They have Windows desktops, iPads and a few Linux netbooks. They have it pretty good.

And this is a London primary school with an Ofsted "needs improvement" rating.

It's not all that bad.

Not particularly excited. After spending a number of years with embedded systems, my tolerance has faded a little. We tried the "follow the instructions" and no help thing to start with. Unfortunately we were blessed with only a composite video cable which is to be fair, a flipping nightmare of reading inconclusive LED states and fumbling around in the dark with no video output. That is where the documentation stops and years of prior experience of embedded systems and knowing what to search for kicks in. Each step came with its own pile of crud to deal with. Each hoop jumped through chips a little bit of interest away. The inevitable question that gets asked is:

"Dad: is this what your job is about?"

It's not and never has been.

The perfect Apple 2 / C64 (or in my case, BBC Model B) was the one that you opened the box, plugged it in and it worked the moment you turned it on without fail, every time and never poked you in the eye unless you told it to.

I still think the (partial inspiration for the Pi) i.e. the BBC Micro is still a better starting point than the Pi itself. Well documented, relatively simple, very powerful, forgiving and the ability to write high level (basic) or low level code (assembly) from the get go from the books that came with it, can play games on it and if you screw anything up, just restart it.

I agree. The point was slightly sarcastic.

However they're just the foundations on which you teach. Programming languages, tools and software packages are the important bits. Which OS doesn't matter really. Neither does any red vs blue pill selection.

Most of my knowledge is applicable to any platform as it's fairly generic. I avoid specialisation for this reason.

I can sit down on any machine and be productive straight away regardless of if it's open source or a feudal empire. That's what we should be teaching.

I don't wish to hijack this thread but tying the Raspberry Pi to education is terrible if you ask me.

I agreed about the Raspberry Pi being the best device for education right from day one. That was until my father bought my daughter one for her birthday and I ended up being the resident "fix it guru" for it.

The thing teaches you merely how to jump through funny shaped hoops to get something working rather than anything realistic or helpful. Most of it is google-fu and copy and paste. When you do finally get there it's a baron land of absolutely unrealistic, undocumented crud that can't self-serve. Plus it barely works and browns out to start with resulting in USB-hub jiggery-pokery (and that only happens because I actually understand how USB works).

For ref, I have 20 years' of Unix and Linux experience (right down to writing kernel drivers) and it was painful getting it off the ground so I'm not approaching it blind.

Being critical (constructively!) of this results in the RPI forum thread being deleted which in itself an affront and a general recommendation against the things.

Windows should be a part of School Education!

MacOS should be a part of School Education!

It's primitive tribalism. People should learn "computing" as part of a school education as over life things are going to change and they're going to need to adapt.

I've had similar experiences with everything Samsung I've used apart from their SSDs. They just don't seem to put much engineering effort into their products.

Acer knock out some good machines. I miss their Timeline series. I got 12 hours out of one of them once and it was super-cheap.

I thought that when I bought my Bravia EX but it's a pain in the ass. iPlayer is poorly maintained and it barely plays anything on a good day, the thing is impossibly slow for the first minute when you turn it on and it's the fussiest thing I've ever seen when it comes to media playback. The TV bit is pretty good but I didn't really buy it for that as there is literally nothing on TV in the UK. Oh and it randomly decides to inform the watchers that it's going to turns itself off even though every option in it is set to not automatically turn off.

It's no Triniton to make a comparison.

I've had zero experience with other smart TVs so this might be the best user experience out there -- please feel free to confirm this or not as I was thinking about getting rid of it.

They can start by starting again. EL is a stinking shitpile of horribleness and always has been. It's the result of the P&P team who obviously wrote the whole thing in a silo whilst the rest of the industry did something wholly different, and better.

Use of EL in a product is usually reason enough to step away from the company.

This isn't actually all that secret. It's been published for over 4 years before it even managed a test flight.

Having worked in that sector, the interesting shit is always what you haven't heard about.

Politics, politics, politics.

Like the last 50 decades at least...

Not really. They barely support it, it's half integrated into RPM and is impossibly difficult to persuade a lot of stuff to work with it due to system dependencies on certain jars. You end up having to pull a full JDK from somewhere else which isn't supported.

If you look at the CLR there are only two major current versions: 2.0 and 4.0 and perfect legacy compat and wide support across all windows server versions.

Yeah agree. Not going past the T420!

No. I complain that Red Hat support a bastardised version of Java (and Eclipse) that ships with their OS meaning that for the sake of sanity you have to deal with Oracle and RH to get a cohesive platform versus the single vendor. Subtly different.

Well I'm stockpiling T-series ThinkPads :)

That really doesn't surprise me. I was given a brand new Sony Vaio VPC-J1 all in one. Had a niceish 20" 1920x1080 screen and an i5 so decided to try and use it for dev work.

Came stuffed with crapware and Sony-isms from the start. Did a rebuild with clean windows. Couldn't find any drivers for any of the Sony proprietary crap (Sony firmware extension parser, Magic Gate, WTF etc) so left them off as it worked fine.

After about 2 months the external PSU blew up. Sony couldn't provide a replacement under warranty for over 6 weeks so I got a dodgy Chinese PSU from Ali Express (took 2 days to arrive!). About 2 months later it started overheating and turning off randomly. Took it to bits to try and clean it and noticed that half the screws on the stand were no longer screwed in - all the threads were stripped. The stand collapsed shortly after resulting it being crudely propped on a pile of books. Turns out there was no dirt in the cooling ducts at all. Took the HSF off the CPU and there was only thermal paste on half of it. Sorted it with some arctic silver. About a month later the disk blew up. Turns out it was from a known bad batch of WD disks. Stuck a new disk in it. A month later it completely failed.

Total piece of junk. It lives in my cupboard as a reminder never to buy anything Sony and as a source of a mouse and keyboard for my daughter's Raspberry Pi. Even the keyboard and mouse suck.

ThinkPad user before and afterwards. ZERO problems ever.

A million times this. We had a small division of our company using Basecamp for a bit but it got canned after some changes and they went crawling back to Salesforce.

It stinks but it stinks consistently!

You're right. But Windows has CLR which is integrated in Microsoft's support offerings.

There are no supported versions by RedHat other than that which ships with the OS distribution. To get a recent JDK we then have to deal with Oracle as well and the cost goes right up. With the CLR and Microsoft there is one point of call under one contract. Plus no one wants to deal with Oracle.

Hardware isn't upgraded. It's replaced when it fails. It's easier to get hardware off the shelf that is certified for Windows Server (any version!). This is particularly true on the tail end of a product lifecycle.

The OS upgrades in Linux are usually utterly painful (Debian included). If you go with CentOS/RH, you have to do this every 5 years at average due to the API churn in Linux distributions[1]. You need to get your developers on there ASAP. With Service Packs and Java updates, a simple test cycle will suffice as they don't break the API contracts. They promise this and deliver.

[1] The kernel syscall interface is fine but major versions of Apache, glibc and compilers and anything even vaguely related to client-side stuff is a PITA.

Depends on what you mean by data.

Google/Facebook mean social and crudely document shaped data. Wikipedia means data that looks like encylopedia pages.

Microsoft mean ANY data you want to shift with the same languages and tools across all platform form factors which is why people buy into the ecosystem. Even going back to 2004, I could knock out an app (in C++/C#) that worked on the desktop, on the web and on mobile devices. In fact I did and it's still in use today managing private couriers with minimal code changes.

Actually no. We get 13-14 years out of Microsoft offerings which is enough lead time for test, provisioning and an 8-10 year lifecycle. Plus due to overlap we have enough time to test v.next.

With respect to support you get a very narrow configuration of supported software and hardware with RH. Keeping it rolling isn't as easy as it looks. The developer support is awful as well. Documentation is shitty,you're tied to C++/JDK versions that are ancient etc. Also no MSDN, no partner support (which is pretty awesome). My few dealings with RedHat support have left us without a solution.

For ref - platforms: Exchange, Windows Server, IIS, SQL Server (the latter I've got an installation that has been cleanly upgraded since 1996 though TWO versions)

Not suggesting Bing. I'm only suggesting "not Google".

Times change. Google's core business doesn't.

Yes because it is instantly replaceable with nothing more than a meh on the part of the user.

I did the opposite. I was a Unix/netware guy for years. Netware fell off a cliff and so did commercial Unix. Left me with Linux and FreeBSD.

Whilst I rather like FreeBSD, when I have to throw an architecture together that will survive over a decade, Microsoft wins every time. They're the only company which provides certainty. RedHat are close but their support sucks.

using a computer without Google products is something that people just don't really do anymore

This is rather naive. In certain circles yes but by far the majority of people use just their search.

That's what the devil on my shoulder said to me but they're just really too inexperienced to make a call on half of the stuff they do.

Ultimately this is probably because experienced people got better offers elsewhere.

I watch github, I watch their blog, I read interviews, I get the picture.

It's my job to look at dysfunctional teams both from a technology and a process perspective. I've seen teams like this many times before. They are expensive, inefficient and the return is considerably lower than the investment has promised.

Just because the historic approaches are worse doesn't exclude these guys from scrutiny. They'll quite happily piss £40k of dev cost up the wall while other departments are arguing over £200 ultrasounds for cancer patients. Scrutiny must be universal and unforgiving.

That is the primary example I'm referring to recently.

There are many more. I'll write them up at some point.

That's fine but the first iterations were because they fucked the requirements up. Regardless of Agile/not, due diligence when selecting products or deciding to write them is required.