|
Kent Sharkey wrote: don't raise a finger
OK
(I raise 3) .|||.
Can they read between lines?
M.D.V.
If something has a solution... Why do we have to worry about?. If it has no solution... For what reason do we have to worry about?
Help me to understand what I'm saying, and I'll explain it better to you
Rating helpful answers is nice, but saying thanks can be even nicer.
|
|
|
|
|
Kent Sharkey wrote: All those surprised, don't raise a finger
How about two?
,|,, ,,|,
Did you ever see history portrayed as an old man with a wise brow and pulseless heart, weighing all things in the balance of reason?
Is not rather the genius of history like an eternal, imploring maiden, full of fire, with a burning heart and flaming soul, humanly warm and humanly beautiful?
--Zachris Topelius
Training a telescope on one’s own belly button will only reveal lint. You like that? You go right on staring at it. I prefer looking at galaxies.
-- Sarah Hoyt
|
|
|
|
|
There was an earlier wave of fingerprint reading, maybe 20 years ago, years before smartphone readers were available. We had exactly the same kind of debunkers then. Mythbusters did a program on it, and for years people lost faith in it.
It doesn't matter that the stealing of your fingerprint requires a whole lot of effort (you don't meet that many possible intruders at the pub, and anyone who steals your empty beer glass before the waiter picks it up is somewhat suspicious!) The process from picking up the fingerprint to creating a false finger that can be used is rather complex and expensive. The lock to be opened must be available to the intruder - he must steal your smartphone or portable, or break into the house where the big PC is placed. In most cases, the risk is near-epsilon.
In those days, fingerprint readers were separate devices. Nanufacturers added safety features, like thermosensors to distinguish between a live finger and a cold rubber one. Intruders had to make hollow rubber fingers to be filled with lukewarm water. The sensors got sensors for electrical resistance between two points on the fingertip; intruders had to switch to a semiconducting rubber mixture. Software was extended to store all 10 fingers, asking you to present a randomly selected one - few intruders have picked up your left hand pinky! (Some fingerprint login systems demanded left hand fingerprint for privileged accounts, for improved security.)
And so on. The best readers became very resistant to fake fingers. But media had no interest in reporting about this; everyone "knew" that fingerprints don't work, as proven by Mythbusters; there was no need to reconsider that conclusion!
Until Apple started it up again. Who would raise any critical comment to something that Apple promotes? It had no thermosensing, no conductance reading, it lacked a lot of the safety features that the separate readers (the good ones) had. But coming from Apple, it must be good!
I wish that we could get back the old, advanced readers for use on plain PCs, either integrated (as they were for some years on the ThinkPad portables, although I don't think they had any conductivity or thermosensors), or as USB devices. Especially when used in 2FA, where you both must provide e.g. a password and a fingerprint, the security is way beyond what is needed in 99.99% of all systems. You wouldn't get the same security on your smartphone (until they add an advanced reader with all bells and whistles, and 2FA), but few smartphones hold information that needs military grade top secret protection.
|
|
|
|
|
"Thinking of tests as validation, it just bakes all these assumptions about your system into a file you run all the time. It creates stasis." "All headlines are clickbait." (some more than others)
|
|
|
|
|
Oh, I see!
TDD done this guy's way doesn't check outputs based on inputs!
Is that what he's saying? I have no idea what he's saying. It's like reading alphabet soup -- all the letters and words are there, but if you try to find any solid underlying meaning, your spoon just splashes soup onto the nice, clean tablecloth.
He seems to be saying that if you write new tests before you change your implementation, you won't need to write new tests after you change your implementation.
Six, meet half a dozen; half a dozen, six.
What I do know is that evangelists will turn TDD, which should be a functional and reasonably usable methodology, into a morass of "Lets follow multiple patterns, rather than get the job done" unnecessary work.
When thinking of adopting a new methodology, it's always wise to check the evangelism level, and to find out how many exceptional circumstances there are, where you have to do things differently.
Of course, to do things differently to this guy, you have to first figure out what he expects you to do.
I wanna be a eunuchs developer! Pass me a bread knife!
|
|
|
|
|
Agile development implies that a developer can change his code at any time, to return another result, or perform a different operation, as requested by some new requirement. Any test regime is incompatible with agile principles. The outputs of a module is defined by what the module returns right now, after the new updates. Module tests for checking that the outputs are the way the outputs are, are meaningless. TDD, module tests, system tests or integration tests are all incompatible with agile development principles.
|
|
|
|
|
You’re designing a set of libraries, used by millions of developers every day, running crucial applications around the world. You want to improve it and add a new feature or enhanced functionality. You must do this, all the while not breaking the millions of existing applications. Despite the evidence to the contrary, they are.
|
|
|
|
|
Kent Sharkey wrote: Despite the evidence to the contrary, they are. what? breaking the millions of existing applications?
Jokes apart... during .Net Framework was fine, now with all other .Net XXX I have the impression, that it is getting messier with each new release.
But... the frightening thought is... It could be managed by the Windows Update team...
M.D.V.
If something has a solution... Why do we have to worry about?. If it has no solution... For what reason do we have to worry about?
Help me to understand what I'm saying, and I'll explain it better to you
Rating helpful answers is nice, but saying thanks can be even nicer.
|
|
|
|
|
Something about seeing what sticks on walls and scat flinging monkeys?
Did you ever see history portrayed as an old man with a wise brow and pulseless heart, weighing all things in the balance of reason?
Is not rather the genius of history like an eternal, imploring maiden, full of fire, with a burning heart and flaming soul, humanly warm and humanly beautiful?
--Zachris Topelius
Training a telescope on one’s own belly button will only reveal lint. You like that? You go right on staring at it. I prefer looking at galaxies.
-- Sarah Hoyt
|
|
|
|
|
In the good old Win32 API, two completely different strategies were used: Either, when a function Func was extended, we were offered the FuncEx interface. Next time it was extended, it became FuncExEx, and so on. There was at least one case of FuncExExExExEx - five extensions to the original.
The other alternative was to put all arguments into a single struct, the first argument being the size of the struct. Arguments required for new extensions were added to the end of the struct. If you supplied an extended, "new" struct to an old library, only the old members were accessed and the old functionality was unchanged. If you supplied an old, shorter struct, a new library would know not to access the extended parameters. The struct size served as sort of an interface version indicator.
I was never in doubt which of these alternatives I preferred, and have tried to push that principle in projects I have been involved in, but most developers reject it fiercely. If a method requires seven parameters, the seven values should be listed in the call, not hidden in a f*** struct!
(Years ago, I worked for a company that had to expedite the release of the new compiler to satisfy an important customer, who had reached the limit of function 99 parameters! The new compiler allowed 256, satisfying the customer's needs. But this was in the days of Fortran, so using structs was no option.)
|
|
|
|
|
Microsoft says IPE (Integrity Policy Enforcement) was designed for immutable and embedded systems (e.g. network firewall device in a data center). What's next? The Linux core team submitting a fix for Excel?
|
|
|
|
|
Kent Sharkey wrote: The Linux core team submitting a fix for Excel? Wouldn't that be a joy?
Click a cell, and a terminal window opens for input.
I wanna be a eunuchs developer! Pass me a bread knife!
|
|
|
|
|
Kent Sharkey wrote: What's next? The Linux core team submitting a fix for Excel? As long as they don't do it in VBA...
M.D.V.
If something has a solution... Why do we have to worry about?. If it has no solution... For what reason do we have to worry about?
Help me to understand what I'm saying, and I'll explain it better to you
Rating helpful answers is nice, but saying thanks can be even nicer.
|
|
|
|
|
Domain experts called corp.com dangerous because years of testing showed whoever wields it would have access to an unending stream of passwords, email and other sensitive data from hundreds of thousands of Microsoft Windows PCs at major companies around the globe. "You break it, you buy it"
|
|
|
|
|
I don't think I've ever seen the word "cr@p" typoed so many times in one place.
Must be dyslexia.
I wanna be a eunuchs developer! Pass me a bread knife!
|
|
|
|
|
And he has been nice just waiting for M$ to come and buy him the domain, without having used any of the data he has been receiving over the years?
Yeah, right...
M.D.V.
If something has a solution... Why do we have to worry about?. If it has no solution... For what reason do we have to worry about?
Help me to understand what I'm saying, and I'll explain it better to you
Rating helpful answers is nice, but saying thanks can be even nicer.
modified 9-Apr-20 9:55am.
|
|
|
|
|
Defamatory accusations aside, I rather suspect that the guy who owned the domain was in fact honest. He could have and probably would have sold to other interested parties had he not been.
|
|
|
|
|
markrlondon wrote: Defamatory accusations aside My error... I should have used the joke icon. Corrected. Thanks for the heads up.
M.D.V.
If something has a solution... Why do we have to worry about?. If it has no solution... For what reason do we have to worry about?
Help me to understand what I'm saying, and I'll explain it better to you
Rating helpful answers is nice, but saying thanks can be even nicer.
|
|
|
|
|
|
Good idea, let me try to get localhost.com for later sale.
Oh sanctissimi Wilhelmus, Theodorus, et Fredericus!
|
|
|
|
|
So many products promise to be machine learning or AI when they are just an impressive algorithm. But smart is not the same as intelligent when it comes to the minds of machines. Can you spot the difference? I'm guessing it's a longer list than the ones that are
|
|
|
|
|
How do you say "plus ça change" in AIML?
I wanna be a eunuchs developer! Pass me a bread knife!
|
|
|
|
|
I'm sure someone could accept a grant to train a GAN to figure it out
TTFN - Kent
|
|
|
|
|
Kent Sharkey wrote: I'm sure someone could accept a grant to train a GAN to figure it out I still say they should have called them GAMs, not GANs.
For good reason, mind[^].
I wanna be a eunuchs developer! Pass me a bread knife!
|
|
|
|
|
A spin-off of “Man about the house” that sounds like “Keeping up appearances”? Sounds like the Beeb has been using madlibs to select shows, not an AI.
How did that show last five seasons?
TTFN - Kent
|
|
|
|
|