matt4711

Say you want to translate a sentence from language A to language B. You have a system that generates 10 possible translations in language B. Now you use a language model of language B to figure out which is the best translation by asking the language model which of the translations has the highest probability of being generated by the model.

There was a post on HN a while ago where somebody wrote a server side tool which can detect curl -> bash piping and delivered different content if that happens.

What about AVM products such as a Fritzbox (https://en.wikipedia.org/wiki/Fritz!Box) ? Primarily a german product but also available internationally (e.g. Australia)

Interestingly, there is a discussion in comments of this paper claiming the authors got it wrong.

But as you said, under the assumption that the test perfectly detects, the results are correct.

Indeed. I was never aware that he was working on his own text editor.

Too bad there are no minimum size restrictions and all apartments in those 30-story towers are essentially shoe boxes.

Lots of Australian ISPs use this model. For example: TPG Broadband Off-Net 50GB -> 10GB peak / 40GB off peak (2am-8am)

We are working on VS support. Last time we tried to port to VS some of the constexpr stuff we use was not available yet. The next release (in 1-2 month) should have full VS2015 support.

Thanks for clarifying this.

The library is very template heavy and thus most of the code is in the header files so I guess it wouldn't change much in this case?

Additionally, we (the main authors) have lots of contributions from other researchers which we would ask for permission to the change the license (I think?).

Most IBM employees in Australia use Toshiba laptops instead of Lenevo.

What is your opinion on this LZ scheme: http://zola.di.unipi.it/rossano/wp-content/papercite-data/pd... which seems to have very good compression effectiveness and decompression time.

besides the sony z1 compact which is actually quite nice. there will probably be a z2 compact soon.

I think he is probably referring to these discussions:

- https://github.com/irungentoo/toxcore/issues/121

- https://github.com/irungentoo/toxcore/issues/578

where the devs are quite defensive about their choices to (1) reinvent the wheel (2) use technologies that are (according to the comments) not secure (3) not audit their proposed crypto/security architecture by experts.

I'm one of the authors of a more complete c++11 succinct data structure library [1] which includes several engineered implementations of wavelet trees, compressed suffix arrays and compressed suffix trees. while you are right that random access is one of the main bottlenecks of succinct data structures, we found that sometimes SDS can be faster than their uncompressed counterparts [2]

The github URL contains some additional resources such as tutorial slides and cheatsheets if you want to check it out.

[1] https://github.com/simongog/sdsl-lite/ [2] http://people.eng.unimelb.edu.au/sgog/optimized.pdf

if they can survive 65000 years on the island I'm sure they have witnessed a few natural disasters and managed to survive as a culture.

the mainframe / high end server market generates only a small portion of the overall profit at IBM. And it has by far not the highest margin (that would be software and services). Just read the earnings report from last year.

are there any technical information available how this maglev train differs from the one operating in shanghai?

looks like this does not support concurrent lock-free operations.

I imagine a compressed suffix array or compressed suffix tree using libsdsl (https://github.com/simongog/sdsl) would probably be a lot smaller/space efficient and faster.

do you have a description of the ".mov tree of structs format"

white girl backpacking by herself through china while obviously not being prepared for what she is going to see there.

not very surprising if you ask me.

her other posts are also very funny in my opinion :p

Flight search engines don't include budget carriers. Most flightfox contests are won by a combination of low cost carriers like airasia or airberlin.

Was looking for some translation guarantees or quality measures but could not find any.

Also the terms of service talk only about "videos" and there are things like

"LICENSE TO OTHER USERS: You further permit other users of the Sovee Service permission to view your videos for their personal, non-commercial purposes, including the right to copy and make derivative works from the videos to the extent necessary to view the videos. The foregoing licenses are in addition to any license you may decide to grant."

looks like this service not ready for any real world translation work.

I'm seeing airberlin (flying to europe) and airasia (flying within asia) quite a lot in these flightfox contests.

I would say programmers using go don't really care how amateurish the mascot looks (see the linux penguin) or how good the typography of the website is.

I would assume programmers care about things like:

- How can I get to the documentation fast - How can I start playing around with the language - What are the language features - How do I install the language and how do I use it

all these things are easily accessible through the golang website in my opinion.

there are so many established android players already. I doubt that would help them much.

At least the premium W8 strategy differentiates them from samsung and co.

So which of these "predicted" events actually happened?

and has been done a long time before that all over europe...