chrisjsmith

IMHO, abstractions become apparent if needed. Abstractions should never be thought of first.

Read GEB - covers it indirectly ( http://en.wikipedia.org/wiki/G%C3%B6del%2C_Escher%2C_Bach )

No - it's simply too big. I enjoy having a voice.

Arrogant idiots! Nothing more can be said.

Likewise. It's an excellent piece of rethinking.

Simplicity equals elegance in my mind. If I can fit the whole product/platform in my mind, I am again powerful and not just a victim of the tyranny of over-abstraction and overwhelming APIs (yes I'm looking at you Java and C#).

rephrase - un-necessary abstration.

Whilst I appreciate the effort that people put into these things, I really don't like this sudden obsession of adding another layer of abstraction over everything. Abstrations are hard to debug, require a learning curve of 2x the original problem and are rarely complete.

You can have plenty of entrepreneurial fun from any age with a bit of street smart without selling lemonade, which would get you a kicking here in the UK.

I've managed to in the last 20 years, apart from my regular grind: clear out schools of old equipment, broker Sun kit on ebay, sell land rover bits, do structured cabling, fix people's laptops, build electronic kits for a fee, dig holes, build sheds, unblock drains, refit DIY stores, distribute fireworks, flatten boxes, skip diving + ebay, selling fruit picked off scrub land, destructive furniture testing, car washing, copper recycling and returning trolleys to a supermarket. Plus probably more things I've forgotten.

I've worked at two successful startups so far and TBH they weren't as fun as most of the above things.

So true. I am an avid TeX user btw. I'm actually addicted to the appearance of Knuth's Computer Modern fonts really though.

Isn't this a step closer to Greenspuns Tenth Rule?

"Any sufficiently complicated C or Fortran program contains an ad-hoc, informally-specified, bug-ridden, slow implementation of half of CommonLisp"

(bar the environment/scope issue it avoids that is)

I'm not a fan of any component in particular, well apart from one surprise, so I thought I'd have a rant:

Word - separation of content and layout is an absolute bugger to manage. Hard to share and version documents. It's actually hard to write a letter or lengthy document with it if you don't have a template to hand.

Excel - utter dog if you need to use it for anything more than trivial and all it does is "ugly" and "puke". most problems are solvable in a cleaner way with a small script (python wins for me). Some crazy organisations use it as a database and email changes around to each other via outlook, leading to sync hell. Sometimes, that is the entire company's business floating around in emails. It's just the wrong tool for most jobs it is given. I usually get the job of replacing these turds.

Publisher - I've seen it used for paper aeroplanes only so I can't comment on its usefulness, or perhaps that is its primary use.

Powerpoint - Many an hour I've toiled in front of the output of this program. HOWEVER, it's far better than most of the awful hour long diatribes you have to sit through with whinging monotonic voices in the form of streaming (or should that be steaming) videos. 2 minutes content in an HOUR of my life and they don't skip properly. At least I can scan through a powerpoint in a couple of minutes.

InfoPath - never used it. It goes near Sharepoint so it's off limit for me.

Access - Access is actually really damn good, but only if you use ADB projects attached to a SQL Server database. If you use local MDB files and Jet, it's another Excel.

Outlook - It's better than Lotus Notes. That's the only credit I can give it.

Office 365 is the above products, done half as well, with a third of the performance, without the old fashioned data security of being able to get it all back easily from a USB stick.

Turned on everything it said would reduce consumption.

Unless you roll a thousand 6's.

That's a fair point. The main problem with that is that most vendors want platform lock in so they implement their own verified loaders (if it's for a phone).

If we had OpenBoot on ARM, things would rock.

Good work Microsoft for taking things seriously.

Thank you for articulating my point better than me.

No the problem is that ACPI is a poor piece of design to abstract what is effectively a stinking great turd of an architecture.

You just simply don't get these problems on non-x86 systems. (I speak from experience with MIPS, ARM and SPARC btw).

It's all down to climate science being more akin to religion that science.

Science is based on building a hypothesis, building a supporting model that produces a reliable theory, then wrinse and repeat.

Climate science is based on building a hypothesis, marketing it, claiming that a theory is going to kill us all, fucking up the numbers when asked to prove it and then sticking your fingers in your ears and ridiculing your peers when a competing hypothesis appears.

There are models for the weather systems on this planet. They simply don't work. We can't predict the weather very well and never have been able to. That is the same for just about everything that climate science has come up with. Back in the 70s and 80s we were facing an ice age. Now it's back to global warming.

We just don't understand why the planet works like it does. We like to pretend we do and make a big kerfaffle about how advanced we are as a species and how we can control our fate, yet it's a thing much larger than the lifespan of the entire human race and it has no respect for us.

I don't believe in god - why would I? It's not observable. Atoms, internal combusion engines and evolution are all observable and we have models which are recreatable.

I think the carbon crisis is political rubbish that spawned a new industry I mean who the hell thought of carbon trading - sounds like something school children came up with.

The earth's atmosphere and environment is more changeable than we realise - we've only been looking at it for a blink of an eye in the scale of things so we can't claim to understand it. Not only that, they've worked out based on various factors that the CO2 levels were higher in the Cambrian, Ordovician and Jurassic periods. It's like trying to work out who farted in a lift full of people.

As for the locavore thing, it's only better because it's not usually mass produced crap so it has health benefits. Chickens, goats etc are a counterpoint to the OP's "theory" as they live off waste and are highly efficient meat generators. Not only that, anyone can manage them.

I grow huge amounts of edible things myself (because it tastes better and is cost, health and psycologically beneficial). I'm not labelling myself with Carbon Crusader, Green warrior, Locavore or whatever terms they think up next etc though - that's just silly.

You're ultimately trading your health for carbon guilt.

Without wishing to trash their efforts, Linux power management is a mess. Nothing works out of the box and you have to play around with powertop to optimise your kernel configuration to get even reasonable battery life. Add to that, the ACPI implementation is buggy as anything.

This, at least in my case when testing on an Acer 4820T, manifests itself as Windows 7 lasting 6-7 hours on battery (!) with heavy Visual Studio 2010 usage, whereas Ubuntu IDLING lasts about 2.5 hours out of the box and 3.5 after tuning it to bits.

It's ok optimising Linux like this but the whole EXPERIENCE is still horrid.

That's because everyone justifiably hates BT, rather than it made sense.

Spot on!

Either that or pi is a symbol and we use symbolic algebra. Just convert to the result in floating point at the last minute.

Ironically my calculator quite happily manages rational form including PI as a constant:

http://h41111.www4.hp.com/calculators/uk/en/scientific/smart...

And still a cunt.

Keeping everything in rational form would be nice, and easy to present.

This is true at the moment but outfits such as Agilent and BK have been making inroads with cheaper, feature-packed kit and agressive marketing. I'd rather have a Fluke TBH. A Fluke won't kill you.

This is a valid response if you ask me.

There is a culture of half-arsedness with some businesses where they don't respect user's security and privacy requirements. This is partially down to plain old incompetance but in my experience it's usually down to the fact that if doing something properly and testing it properly doesn't add business value, then it's not done. At the risk of pissing people off here; that culture is prevalent amongst startups.

They screwed up, they're getting sued. They should have tested it properly.

If this was a public organisation that left everyone's files in an open skip overnight they'd get sued too.

The lack of errors was annoying and counterintuitive. I like the usual UNIX mantra of blow up noisily if an assumption is made but it just doesn't happen there.

Slightly OT (or is it), but it's an ad for Fluke - probably the best test gear you can get IMHO yet under-advertised. It's good to see a good company getting with the times.

Been there done that. It's very annoying indeed. I wish things were slightly less finicky in *NIX platforms as a rule. They need a serious de-crufting session (plan 9 looks very much less crufty).

choke SICP.

And it works in bright sunlight and doesn't require batteries.

On a more serious note, Khan Academy is great but the format allows too many distractions. I'd like to see a paper version of it.

You might be dead before you get to spend it.