Catch up on stories from the past week (and beyond) at the Slashdot story archive


Forgot your password?

Comment: Re:Google Picasa 3 (Score 1) 259

by famebait (#48599277) Attached to: Ask Slashdot: Best Software For Image Organization?

My only quibble with Picasa is that it the Mac version doesn't have proper retina support yet (which you REALLY want in a photo app). I'm sure it will be fixed eventually.

But otherwise Picasa is fantastic. Very good at using the actual files and filesystem rather than a proprietary database.

Whatever you chose, make sure it encodes metadata IN the actual file or filename, so you can shove a subset to whatever other app or platform you like and it will tag along with no process required. Given the choice, in-file is better than filname, so that sync tools can match up different revisions of the same image.

+ - Poll: Do you worry about the singularity?

Submitted by famebait
famebait writes: - No, it's bollocks
- It is inevitable eventually, but so are lots of things
- Sure, I expect it to become a real problem in my lifetime
- It is already happening
- No, it'll be fun.
- The singularity should worry about me

Comment: Re:Almost makes me want to live there (Score 2) 77

by famebait (#46693821) Attached to: European Court of Justice Strikes Down Data Retention Law

Nope, the EU 'government' created the damn thing in the first place. The EU *court* struck it down.

OTOH, at least the whole scheme was out in the open, as opposed to the plain illegal surveillance in the US.

OTTH, who knows how many of the national goverrments are doing that as well.
There is also the neat trick (used by the UK among others) of turning a blind eye to, say, US surveillance of european citizens in return for access to the data. That way noone is offiically spying on their own citizens, but the net result is the same.

Comment: Re:Minimise state (Score 1) 373

by famebait (#46600749) Attached to: Ask Slashdot: What Do You Consider Elegant Code?

No, I love objects.
But language features should be used where they bring value, not randomly because they're there.

First of all, I have no problem with immutable stuff, I don't even consider that state.

Mutable object state is fine too, as long as it in fact reflects the behaviour you're modelling. But it should be pared down to that. Not absolutely religiously, but it is a good ideal to strive towards, right up there with high cohesion and low coupling.

Many litter their objects with instance variables that do *not* model the core state of the object at all, but rather serve as a 'nifty' way to pass data to functions with less typing. This makes the object more stateful, which just makes life more difficult all around: you need to know what has happened before to predict what a function will do, and so you depend on many more places in the code to be certain that any one piece does what it should. The logic becomes less readable, more bug prone, harder to alter without introducing bugs, and harder to verify with tests. And that's not even going into thread safety and parallelization.

Fighting all this is not fighting object orientation. If anything, some solution approaches can be considered *more* object oriented: if that param-list starts getting too long, rather than abuse instance fields as lazy params, group the params into sensible value objects, and you may find the new abstraction yields additional benefits: an avenue for better expressing function and intent, implementing validation and constraints, encasulating calculations and query logic from the client code where it only constitutes noise, into self-explanatory functions on the data object.

Comment: Re:Minimise state (Score 2) 373

by famebait (#46583107) Attached to: Ask Slashdot: What Do You Consider Elegant Code?

Caching is an optimisation that does indeed introduce state, and all the problems that comes with it in full.

Like all optimisation, it constitutes buying performance in exchange for increased complexity (of some sort). The cost of complexity is high, and the tradeoff should be qualified by a solid cost-benefit anaylis, and by extension pertain to a demonstrably real performance problem.

That said, I certainly accept that 'true' changing state can be part of a good model, and I don't consider that caching. Just keep it under control, make sure it actually delivers value, and don't sprinkle it around lightly.

Comment: Re:Comment your damn code (Score 1) 373

by famebait (#46582589) Attached to: Ask Slashdot: What Do You Consider Elegant Code?

I was intentionally overstating it a bit.

There are of course cases where comments are warranted, and properly justified optimizations may be just such a case. The others also usually fall under the broader umbrella of "exceptions" to how things might normaly be done. Just don't use it as an excuse to make the code less clear than it could be given the perfomance constraints, and beware of premature optimisation, which is a prime cause of brittle and unreadable code, with frequently no real benefit to offset the cost.
And always consider if you could say some of it with code too.

As a general rule, though, I still think it is wise to keep in mind that since the comment is not executed, there is no guarantee that it reamains correct, if ever it was.
I sometimes make a hobby out of trying to find at least one error in every comment I see. It doesn't always pan out, but tre percentagewhere it does is both staggering and frightening - I warmly reccommend the practice to everyone.

Comment: Minimise state (Score 5, Interesting) 373

by famebait (#46582327) Attached to: Ask Slashdot: What Do You Consider Elegant Code?

You always have to keep in mind that code will be changed by serveal peopale, and your 'elegant' intention may not be understood or followed through by the next guy. So go for simple rules of thumb that not only keep your code readable and clear, but can accommodate future change while ramaining so.

My number one rule for keeping code both readable and robust is this: Reduce state.

I don't mean everything needs to be purely functional, but consider state a general liability to both correctness, readability, testability and maintainability. Less is more..
* Whatever state you have should be focused and serve to explain/model the actual problem domain, not just 'keep stuff for later'.
* Keep state as local as possible - most code is litered with instance variables that should have been locals and params.
* Just because an object _can_ bundle state with its functions doesn't mean it _should_.
* If it can be done in a static method and still make sense, do so.

Comment: Re:But it is horribly wrong anyway. (Score 1) 458

by famebait (#46078855) Attached to: Stephen Hawking: 'There Are No Black Holes'

Who said anything about GPS? There there is plenty of stuff in geosynchronous orbit that need a clock accuracy that requires taking relativity into account, and has been from way before GPS, regardless of where the first experiments to demonstrate the effect took place.

In any case, proof at NEO invalidates the newtonian prediction, and I know of no model that predicts that the problem would exist at NEO but go away at higher altitudes.

Comment: Re:But it is horribly wrong anyway. (Score 1) 458

by famebait (#46056705) Attached to: Stephen Hawking: 'There Are No Black Holes'

You're pretty sure. Oh, that's settled then. Don't bother googling it or anything, it's not like anyone actullay knows this stuff or publishes anything about it.

FYI: Merely the altitude of, say, geostationary orbit implies a potential energy that means you have to account for time dilation if you want to stay in sync with clocks on the ground. This was proven experimentally decades ago, and predicted way before that.

The other line moves faster.