|
The actual paper[^] (why don't we get those in the insider news btw? it's not like we're laypeople who need the layman explanation) mentions they use MS-SSIM and PSNR-HVS as quality measures. Certainly a better idea than using PSNR or (god forbid) MSE, but still blurred too much. It seems that everything underestimates the importance of visual energy and over-penalizes noise, even though the point of SSIM was to compensate for that. Noise isn't necessarily so bad, it adds visual energy which we are more sensitive to than the exact phase of the higher frequencies, but even SSIM really prefers the phase to be the same too (because it cares too much about the covariance) and therefore ends up penalizing images that look fine but differ by a lot in many pixels because the noise moved a bit.
They should try this with an actual psy opt metric.
|
|
|
|
|
So it's not just Google, the entire industry is elephanted. That makes me feel so much better.
Are we really sure that any of these new fangled modern formats like .bmp or .gif are really better than a simple .pcx (run length encoding FTW!) file?
Did you ever see history portrayed as an old man with a wise brow and pulseless heart, waging 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
|
|
|
|
|
Remember, this is the same company which claims VP9 is better that H.265.
(Google's vaporware is interesting in that it exists, but rarely lives up to its promise. Their engineers seem to get bored at the proof-of-concept stage and walk away, so almost everything they make seems half-baked. And they don't care. But, what if they did care? The world trembles.)
modified 26-Aug-16 15:46pm.
|
|
|
|
|
So true. I think they like to toss stuff out, expecting everyone else to jump in and finish it (ala SPDY -> HTTP/2)
TTFN - Kent
|
|
|
|
|
Oracle has revealed its interim plan to help Java devs deal with browser-makers' imminent banishment of plug-ins. Just in time for when no browser will support them
|
|
|
|
|
Who still using it?? so many security loop hole
|
|
|
|
|
I'd bet there are still a shocking number of mission-critical enterprise apps that only run as Java applets...or as ActiveX controls that only work when embedded in a page in IE4.
|
|
|
|
|
They're just marking it as obsolete, but not indicating that it's going away... So basically anyone stupid enough to only look at whOracle's documentation will see something that only appears to acknowledge the events of the last decade have made java craplets not the preferred way to do anything; but receive no warning that within a year or so no modern browser will be able to run them at all. 
Did you ever see history portrayed as an old man with a wise brow and pulseless heart, waging 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
|
|
|
|
|
A big reason for low adoption could be that they are less secure. And while many are touting the security of biometrics, there are four issues to consider when evaluating the technology. You can't cut off someone's password?
|
|
|
|
|
 agree and again you can not read mind for now
|
|
|
|
|
|
Biometrics are more safe because some of the lazy people which dont use will use it.
And passwords can be recorded by cameras or key loggers.
It all has "some level" of security. 
Press F1 for help or google it.
Greetings from Germany
|
|
|
|
|
But they use it in all the sci-fi movies -- it has to be more secure!!!
Marc
|
|
|
|
|
The million-dollar question about the level of productivity in software development teams is the following: How does their productivity scale with the team size? "Nine women can't make a baby in one month"
|
|
|
|
|
Any big team is going to be slowed down by having to coordinate almost everything. I think on small teams there can be improved productivity if they team members are highly skilled is specific areas, and their skills complement each other. I have definitely seen productivity go down on a lot because of other developers. I worked with a developer that had a tendency to much with my WPF code without telling me, and he was not a front end developer. Was working with Seapine Software's SCS, and it is pretty bad. He insisted on working directly under the integration branch while everyone else was working in the same feature branch. I managed to overwrite his changes and he insisted I back out. That does not work too well with Seapine since it does not handle deleted or added files well. I lost like 5 days, and his changes were minor in the project I was working in. This was the primary branch I was working. Did not have this problem with any of the other developers.
|
|
|
|
|
Actually these days developers scale reasonably* well - once we've got the common architecture, tools and source control issues worked out.
Unfortunately the ambition of the project owner seems to scale exponentially - i.e. everyone seems to think "It should be like Excel but with these 18 additional functions" is somehow sensible.
(* Say 0.3N?)
|
|
|
|
|
haha agree  My Associate manager at Accenture alway told me that
|
|
|
|
|
To be fair, I believe credit is due to Fred Brooks: Brooks’ law - Wikipedia[^]
"If you don't fail at least 90 percent of the time, you're not aiming high enough."
Alan Kay.
|
|
|
|
|
thank you, for point it out
|
|
|
|
|
I think every project has an optimal size. A big team has a lot of man power, but coordination, interfaces and "ramp up" time is getting bigger and bigger. Often occurs the problem one specialist is waiting on other results. (I am often waiting on our fancy design team)
The "smallest team" is one person: Knows all about the project but can only do one job at a time.
The optimal size comes from how much work can get done in paralell. 
Press F1 for help or google it.
Greetings from Germany
|
|
|
|
|
Three to five.
Almost any large project can be broken down to discrete projects of 3-5 engineers. (When you add in testers, documentation and perhaps a build-engineer and manager you get about 8-10 people, which is about the size of a military squad/section.)
|
|
|
|
|
Researchers have devised a new attack that can decrypt secret session cookies from about 1 percent of the Internet's HTTPS traffic and could affect about 600 of the Internet's most visited sites, including nasdaq.com, walmart.com, match.com, and ebay.in. "It works in their laboratory". The new security model.
|
|
|
|
|
In a world that's growing ever more reliant on apps, the rapid pace at which they're being developed and deployed is causing significant problems with security, according to a new report. The latest study from the journal, "Duh!"
|
|
|
|
|
Gee, who'd'a thunk it?
#SupportHeForShe
Government can give you nothing but what it takes from somebody else. A government big enough to give you everything you want is big enough to take everything you've got, including your freedom.-Ezra Taft Benson
You must accept 1 of 2 basic premises: Either we are alone in the universe or we are not alone. Either way, the implications are staggering!-Wernher von Braun
|
|
|
|
|
The big question is whether Proxima b has an atmosphere or is a cold, dead world. I think I played that game a while back
|
|
|
|
General 
News 
Suggestion 
Question 
Bug 
Answer 
Joke 
Praise 
Rant 
Admin 
Submit an article or tip
