Posts by Frumious Bandersnatch

Re: What a horrible waste of time and money

Alternately this is the only way

Surely you mean "alternatively", Shirley? (and yes, I did call you Shirley)

Re: And they would have to sell the chickens to buy the food to feed them.

This also reminds me of one of the contributory factors in the great potato famines in Ireland. Smallholders lived off a subsistence diet of spuds while cash crops like grain were by and large exported.

Granted, in this case, the cash crop (chickens) are owned by the small farmers themselves rather than the landlords, but if your subsistence farming isn't going so well, those chickens are going to start looking mighty tasty. I won't be so churlish to point out the supply/demand side of things if suddenly everyone is selling chickens... (ok, I mentioned it)

It's a noble gesture at least, but I think you need to need to tackle both aspects (getting better/more reliable yields from subsistence farming and cash crops) at once.

Plus, how much research has gone into the particular breed of chicken being given out? I would hope that there's a pretty diverse selection (good, wide genetic pool) with particularly hardy breeds suited to the local conditions.

Re: "never fails to amaze and appall"

This is a murcanism that I was not aware of. My apollogies.

(Incidentally, it has helped me fulfill my downvote quota for the month. Many thanks)

@ John Tserkezis

Your objection has been unduly noted.

Re: A Monster Calls

Is that the kids' book about the lens-grinder from Omsk?

"Plus one has to calculate the current angle of the sun [etc.]"

Why? Every schoolboy can figure out how to steer the sun's reflection in a wristwatch so as to dazzle somebody. No calculations required.

All Archimedes' rig would need (I'm speculating) would be a smaller targeting mirror with a shorter focal length attached to the main mirror, along with a separate targeting reticule (I think it's called). Then targeting would just involve moving the rig until you have a line of sight from behind the targeting mirror, through the reticule (which would be lit up) to the ship beyond.

The real problem, as you mention, is the quality of the mirrors and their fixed focal length. You would need a fairly large number of these to set fire to a sail.

Re: Functions returning multiple values.

Perl subs (and some builtins) have this too, via the ability to return a list. You can aid readability firstly by properly documenting the calling convention, but also by using constants to simulate enums. For example:

Of course, Perl is a pretty pathological language. You can even modify the type of thing returned (via something like wantarray) depending on the calling context. Loads of scope to shoot yourself in the foot.

Re: Hat off. Beer raised.

It turns out I can still remember 6502 assembly op-codes...

I think I can only remember A9 (LDA, immediate mode?) and EA (NOP).

For shits and giggles, I tried to write a Hello World program without looking stuff up. Can't remember exact instructions and addressing modes, but I think it might go something like ...

Re: in a way, but

Ransomware can also permeate into backup media

True, but keeping an eye on the backup process can help detect large deltas.

The way I do backups has been the same for many years:

• Use Linux and ext* file system
• increments start by making a hard-linked (cp -l) copy of previous snapshot
• Use rsync or similar tool that only overwrites/transfers changed files
• Similar arrangement for 2nd, 3rd generation backups

If something were to start encrypting files en masse, I would see it pretty soon, either in the rsync summary (being longer/larger than usual) or in the size of the increment as stored on the disk---after the backup, I calculate the delta size by counting files that only have a single hard link; these must be the changed files. Because hard-linking takes up relatively little space, I maintain these "snapshots" going back for quite a long time and only delete them manually, so that gives me a second chance to notice any damage and to roll back when it does happen.

I also use a hand-rolled file integrity system based on the same idea as the "shatag" tool. I will periodically update SHA256 hashes for all files and store them in the file system as extended attributes. I also collate these hashes across all machines and use the metadata to enforce a replication policy across multiple machines (or at least to verify that it's working). I've also got a separate scheme (using erasure codes to give a high level of redundancy with modest overheads) for cold/archival data.

One other thing I've toyed with is using the LVM snapshot facility. It could replace the hard-linking scheme I use to some degree. In this case, larger-than-expected deltas would overflow the copy-on-write buffer, alerting me to something strange/unusual via a message about a failed backup. I prefer the hard-linking scheme, though, since it's more permanent and gives better historical integrity. LVM's snapshot facility is perfect for backing up volumes with databases on them, though, since you get an atomic backup without needing to lock the database first.

Re: "photos of my nob"

noun (cribbage) ...

But that would be "one for his (ie, the Jack's) nob(s)" (and two for his heels)

Re: RAID5 no longer has a role with hard drives

Therefore you should be using big RAID sets, like 14+2.

But if you're going to be using such a large number of disks, it makes more sense to use an erasure code. I assume that 14 + 2 means that you have 16 disks and you can tolerate 2 failures.You might think that three near-simultaneous failures is going to happen infrequently enough that you can ignore it. I guess you've heard people talking about waiting ages for a bus and then two come at once. It's all based on the Poisson distribution: (independent) rare events can and do happen in clusters. You might say there's more chance of winning the lottery and being struck by lightning but things like that do happen

I found a calculator tool and it told me that in a 16-disk setup (16 x 5Tb), with 10Mb/s available for rebuilding, MTTF of 3 years and resupply time of 7 days (no hot spares), the chance of a data loss is 1/37.6 per year for a RAID 6 array.

A once-in-40-years chance might not sound too bad although that is only for one array. If you're in a data centre with 40 arrays, you can expect around one such failure per year.

Anyway, to actually get to the point, you use the Poisson distribution to calculate the likelihood that a certain number of independent disk failures won't happen in the window when you're rebuilding the system. The more disks you add, the higher the probability that these rare coincidences will happen. The best mitigation for this is to increase the redundancy level, so that if instead of a 14 +2 scheme, you used a 12 + 4 one (ie, an erasure code) you're (roughly) exponentially less likely to suffer from catastrophic failures.

Add in the fact that Poisson arrival rates are only an assumption, and that clusters of disk drive failures can happen more frequently than the model suggests (eg, bad batches from a single manufacturer), it makes even more sense to use an erasure code for arrays with many more disks than the standard raid setups (more than 4-8 disks).

Re: Pah!

MMmmm. I like Ruth Negga.

The Undertones: https://www.youtube.com/watch?v=tSdsTkqerOw

Happens all the time /

Its going to happen - happen - till your change your mind

"Firstly, cognitively you can't keep 200 items in the stack"

(TL;DR at the end)

I'm guilty of keeping hundreds of pages open at a time. Right now I have 4 windows with 214, 180, 158 and 102 tabs. In my defence, I'll say first that it's not a stack. It's more like a serialised/flattened tree (or actually, a forest). When I middle-click a link, the new tab opens next to the referring page so the flattened tree structure is maintained no matter how many pages I open.

Most of the pages that I have open relate to some particular search topic that I've been interested in following up on. The easiest way to do that is to speculatively click on a bunch of promising-looking results from a search engine, scan some pages and then either refine the search or drill deeper within existing search results or the sites that I've already opened.

I don't think that using bookmarks is a very good way for dealing with this kind of ephemeral collection of pages, although if the browser had a feature to bookmark (or pop out into a separate windows) a range of tabs, I would definitely use that. Instead, If I want to jump back into a particular search tree, I just go to the address bar and type some relevant keyword and use the "switch to tab" feature to find where I was when I went off to do something else. This is much easier and less work than using bookmarks or trawling through the history (which is basically unusable in Firefox) to try to recover the state of my search trees.

Every so often I do a sweep of open pages, starting from the most recent (rightmost) tabs. It's usually easy to spot a range of tabs and delete them all (individually; again, a "delete range" function would be brilliant) without needing to scan the contents. If I remember that there was something in that tree that I might want to come back to again, I'll find the best links and note them in some way (in a bookmark folder or in a wiki that I use for note-taking) and then close all the tabs. If I know that I haven't finished some search, I'll skip that range and deal with it in a second sweep.

So anyway, the TL;DR: if you have enough RAM to be able to keep loads of tabs open, it makes for a very easy and lazy way of keeping on top of tons of disparate islands or pockets of information that you're interested in. You probably want to scale back on doing this sort of thing on a work machine (find some other way of reminding yourself of things to check out later) but it's nice in the comfort of your own home.

Re: If Only...

And what the heck does "our learnings" mean, anyway?

It is an Internet Servings Providings company, after all.

(and no, just because gerunding (oo-er, missis) verbs can be done, it doesn't mean you should)

Re: Pedant

I missed my 10min edit window ...

Thinking about it a bit more, I think "fount" is more like a "wellspring", while "font" is more like a "repository". I think of wisdom as something more gained and accrued* than being a natural upwelling so a "font of wisdom" sounds more natural to me.

* “Good judgment [wisdom] comes from experience, and experience comes from bad judgment.”

Re: so....

On *nix systems it should be easy to write a one-liner to search for all executables and use ldd to check what dynamically loaded libraries it depends on. I assume there's something similar to ldd on Windows. Problem with that is, I guess that people who develop Windows programs tend more towards using static linking. Maybe EXE explorer [link] can help there. The stackoverflow thread here suggests using the DUMPBIN program that comes with Visual Studio.

Re: Nonsense

So how exactly do you do a magnetohydrodynamic drive "properly", then? Turn the air in front into a plasma and eject it with a magnetic field (Lorentz, not Laplace, I assume)? Sounds totally impractical to me.

Re: Sounds like geckos' feet

Gecko feet exploit the van der Waals force. As far as I can tell this is different from an electrostatic force as mentioned in the article. This looks to be more like how vinyl "L" plates for sticking inside car windows can stick to the glass without adhesive. I *think* that's an example of electrostatic attraction, at any rate.

Re: At the BBC, Agile means 'making it up as we go along'

re: Give me a waterfall any day

I'd rather have something more like Barry Boehm's Spiral model. Make risk management part of the culture and try to tackle the biggest risks first. Like maybe, you know, in this case they could have thought "let's try to pin down requirements because right now we really don't have a clue."

In theory, I guess Agile is supposed to be some sort of successor to the Spiral model, but as someone else mentioned, it just ends up with lots of little waterfalls and no overriding sense of direction. It seems that some sort of magic is supposed to happen because "tools" or "teams" or "continuous delivery" or whatever. So much (under)pants (gnomes), I say!

Re: Absolutely horrendous...

no Ortolan ...

but sounds quite like cuy chactado (squashed and fried guinea pig, as already featured as a post-pub deathnosh).

Re: Why the fuss?

Call me blasé, but what is all the hype with this event?

Maybe astronomers have a soft spot for Mercury transits since they were a key support for Einstein's theory of relativity? The thoughts of overthrowing the older Newtonian hegemony probably makes them a bit nostalgic.

https://en.wikipedia.org/wiki/Tests_of_general_relativity#Perihelion_precession_of_Mercury

The only way that could be worse is if a huge manatee was involved.

Re: Alice...

I think she had a restaurant ...

Re: Can someone explain . . .

The magic of geometry

And some even simpler linear algebra. It's like* if you have two buses that serve the same bus stop, one that arrives every 40 minutes, another every 45 minutes. The time between instances where both buses arrive at once is the least common multiple of the two times, which in this case would be 360 minutes. Accounting for the wobble is like saying that you only visit the bus stop every, say, 50 minutes so you're only interested in times that you're actually there. Again, you use the LCM. The LCM of 360 and 50 (or of 40, 45 and 50, if you want to combine all three values at once) is 1800, so the time between the coincidences is 1800 minutes or 30 hours.

* Obviously, this is a simplification. The bus stops would be moving, for one thing, since we're interested in colinearity rather than when planets are at fixed points. In two or three dimensions with elliptical orbits, the calculations are a bit more involved, but the basic ideas of periodicity still hold (as far as I know; please correct me if I'm wrong). The reason I'm talking about the simpler case is that it helps to understand that the LCM is fundamental to combining periods. Most notably, if the periods being combined are relatively prime, then the combined period is the product of each of the individual periods, which might be a surprising result if you didn't know about the LCM.

Incidentally, they reckon that cicadas are so successful because the period of their life cycle is relatively prime to that of the predators that keep them under control. This means that they get the maximal period between "busts" in their predator-prey cycle.

Re: That's the unix way of doing things..

I'll tell you a true story. Back in Uni, we had a practical programming exam (in Basic) on the mainframe. The lecturer had set up a restricted environment where commands that could be used to cheat (those relating to sending messages to other users and accessing shared folders) were disabled by using aliases. I noticed that I could undo these aliases from within the Basic interpreter. I hacked the system by asking the lecturer if we could use the Basic interpreter during the exam, because it was more convenient for testing things quickly. They didn't see the problem and whitelisted the interpreter. So after finishing my assignment, I had a bit of fun messaging my mates to show that I'd broken out of the jail.

The moral of the story is not that there's anything wrong with interpreters (like your diatribe against shells) but the context that they're allowed to be used from. ImageMagick evolved from being a command-line tool and now it's being used in an unsafe context. That is all.

Re: Fusion?

Thanks, that saved me the bother of asking "wft ...?"