Re: Prevent fraud ?
So you don't understand machine learning, and can't be bothered to Google it, but want someone to summarise the state of the art?
This technique has been used for ages in reCaptcha - which is (or was) used by ... Amazon.
40 posts • joined 30 Mar 2012
Obvious quote - "You're only supposed to blow the bloody doors off!"
Thermic lancing it would undoubtably work quickly, and almost certainly destroy the contents so kind of pointless.
As they said in the article, the safe contains "sensitive equipment" - not much point in beating up the safe if it destroys the hardware token in the process.
Intel produce products - I'm sure there are some cases where they've gone after competitors for patent infringement, probably some of them unfair or anti-competitive, but at least they're using the patents they have in things people buy.
That's fundamentally different to a firm which has never produced anything, has no intention of ever producing anything or setting up a reasonable licensing scheme (e.g. like ARM does), and doesn't even benefit the original patent author.
I guess the chances of a hash collision are so much smaller than the collisions we already have with a combination of name + types that it isn't a priority to worry about.
The second point is probably an advantage - no unexpected changes, the references must be _deliberately_ updated; it also allows parallel testing and stuff in a real environment, and you since you can uniquely identify a function you can easily see if the old one is being used.
Wow, could you have put more incorrect statements into one comment?
It is a problem with their website if they are losing orders. The customer may not always be right, but losing an order through a technical problem is always wrong.
Credit cards do frequently have four digit years, and if the site asks for "year" it is perfectly reasonable to put in four digits.
Autofill programs may fail, but it is the dev's problem since that's the platform they're working on. And it's a perfectly testable scenario, so no excuses it's just poor coding.
If you've ever looked at any research on loss of capture through UX issues you would know that the percentages lost can be staggering, even at payment stage. You do not want to give people time to get pissed off, think about if they really want it, or look at the total price again.
I agree they should protect the keys...
But I have no sympathy for someone who thought it was OK to publish any of their company's stuff on GitHub without asking and receiving clear permission.
His employment contract will certainly have had the usual clauses about company ownership of his work, confidentiality etc. Even without the keys he broke his contract and trade secret laws, it's just unfortunate the keys made it a lot worse which is on the company.
The "other wrist" thing in the end comment is actually a point. If you put the watch on so the text is the right way up, the buttons are probably not usable or may act in the wrong direction.
iWatch has a right/left setting to flip the screen, but surprisingly it looks like Google Wear doesn't support this you need a separate app!
Did you decide to write this as a rant, and then ignore all the evidence?
On the link you provided you can filter the data pretty much any way you want, or download it in CSV format to the spreadsheet / database or your choice. The link is even right at the top of the page and craftily labelled Search... obviously they were trying to hide it.
I get the bit about price, but a. that changes all the time and b. recommended prices are for idiots and c. if you have specific requirements, you normally narrow down the search to a bunch of options, then look for prices
The trouble is that if the company goes bust, maybe because of these requirements, nobody wins - and it's already far to easy to just claim bankruptcy to avoid responsibilities.
Mandatory escrow of all supporting code and IP rights would make more sense - any company buying the rights would have to think twice about shutting stuff down if the code then became public. Half of these platforms already have most of the "support" run by some community which could take ownership.
Maybe add in joint ownership / first rights to any relevant hardware, IP addresses, domains etc as well, so there's a viable transition period.
I've bought a Surface, it's great - except MS are spamming me with a "Your Surface" email with nonsense about how I should use it, almost every day.
It has no unsubscribe option, it's plain marketing spam from my viewpoint. It even says "forward to a friend" - no, that is not ever ever going to happen!
Any ideas how to unsubscribe? Nothing in my MS account (I already have "don't send me junk" selected), nothing in my MS profile already removed any subscriptions.
Ditto OneNote but I guess that's another forum... seriously, one new device two new spam subscriptions from MS.
I've tried posting on the MS forums but just got unhelpful "Oh MS would never do that" from a fanboi.
- on insecure irrelevant sites
I use the same password on any number of forums and support sites where I really don't want to spend the time remembering them and a breach would be completely uninteresting to me.
Half the sites probably run old or homegrown forum software and dump the password straight into the DB in plaintext anyway.
I combine this with a disposable email address in case of a spam overload, but so far they're all so irrelevant that nobody has even bothered to hack them and the address is pristine !
I was thinking the same thing.
For confidential email, why would anyone trust their email provider and not use encrypted email?
Of course, it is virtually impossible to hide the source and destination in email, since that's needed for delivery and having a two way conversation, but even that is fixed by encrypted online form - just use it from a web cafe, and provide a unique private ID to have a secure two-way discussion.
Of course she would need to ensure the encrypted data is only opened on her own PC in her own secure environment.
The blog owner can still be subpoenaed for information but that's just normal law exactly as it should be; I'd imagine PJ would be able to handle it, and have lots of support if it happened.
This post has been deleted by a moderator
But the point is also that a team of end-spectrum nerds is frequently not the most productive. Deep technical skills good, ability to apply to real world problems or timelines not so much.
Unless you're a pure technology company, having an IT person who can work with reality is pretty important, and the gender balance in many non-IT functions is equal or at least not so imbalanced.
Also increasing the frame rate vs resolution would be easy to implement in TV hardware - sets already show far more than 25/30 fps, and I happily send 60 fps over HDMI from my PC.
The biggest cost for these new formats is the increased resolution; if they forget that I'd love to get an improved experience on my perfectly adequate current HD TV.
You'd never get the telly makers to go for it, but imagine if one of the big broadcasters picked it up and sold it as an upgrade requiring no new hardware or just a new STB not a new TV.
Could actually be good if the photos are taken by the reporter, at the time, and line up with what is written. How many articles have some stock photo, or the same as all the other articles, and give me a strong suspicion they were paraphrased from someone else's work...
Some of the best articles on this site have photos by the hacks, usually accompanied with something like "it's a crap photo because the light in the pub was poor and I've only an iPhone 4" but at least you know it's real :-)
One obvious differentiator is the resolution - 758x1024 rather than 600x800 - and it isn't commented on at all !
As an owner of a 600x800 E-Ink reader (Sony, fwiw) and a retina iPad, I find it hard to go back to the reader even with the much better battery life and more appropriate size because the text is so jagged.
So reviewer - is the higher resolution any better?
Actually an alpha emitter would be completely useless to mimic cosmic radiation etc. Regular alpha particles don't even pass through skin, so unless you took the lid off your chip and attached the Americium directly to the silicon it wouldn't have any effect.
I'm assuming levering the lid off a chip would cause some problems anyway, but maybe the teardown sites could give it a go :-)
I think the point is that anything you put on the company network / email belongs to the company as per contract and reasonable sense. This seems to be a case where they're trying to get hold of a copy from externally because they deleted their own copy.
Perhaps they could argue that the copy is theirs due to intellectual property rights, same as a photocopy would be, but in that case I'd think he could just delete the mails as it isn't his responsibility to keep them.
Bullshit implies they claimed something untrue. They claim unlimited backup, and throttling the upload speed doesn't make this untrue although you could argue it makes it less useful.
As I said originally, for most home users who have a lots of unchanging data such as video and photos, I don't think it would cause a problem or invalidate the point of their offering.
If you need to store more that 200GB and have a large amount of change then it would - take your pick.
To be fair to Carbonite, they are offering a simple home service and they don't claim unlimited bandwidth. Once the backup is up-to-date it doesn't take much transfer to keep it that way, so the restriction is probably not an issue for the vast majority of users.
I've used it for a few years, because although I could do something more complex, faster etc, if whatever takes out my machine also takes out me then Carbonite is simple enough for my partner to understand to get our stuff back. I've also restored some fairly large video files without issue.
I don't understand why we can't have a system where the patient controls the access.
* Patient carries a smartcard - if they go to a hospital or are found lying in the street, smartcard gives access. Put in some sort of emergency override if necessary.
* Patient registers with doctor/provider - uses smartcard to give access either permanently to that practice, or for a period in case of private one-off things.
* Patient has access to web portal to add/remove these bits, and see who has been accessing their data.
This is kind of similar to the way lots of organisations already work with 2 factor authentication and access to service accounts. Because the data is stored centrally it's a lot more robust in case of card loss, but you could add an option to load it onto the smartcard to help with network outages.
Biting the hand that feeds IT © 1998–2020