Cover photo
Nita Vesa
Attended Erkko-lukio
Lives in Ilmajoki
140 followers|69,190 views


Nita Vesa

Shared publicly  - 
Hmh. I was trying to get audio/video chat working with my XMPP-server and it appears I'd need a STUN/TURN-server, but what little I can find about such with Google-searches is that you need two public IP-addresses for that and, well, I don't have such. I don't have a business-class broadband, I can only get 1 public IP-address. Also, I was a good while googling around and it seems like there's no way for my XMPP-server to tell clients to use some 3rd-party, external STUN/TURN-server. Maybe I'm missing something, but at least I can't find anything like that.

This bullshit annoys me to no end. I just want audio+video+text chat on a server that I control, but it seems even that is too much to ask for.
1's profile photoLuke Tidd's profile photoNita Vesa's profile photo
+Nita Vesa which xmpp server software do you use?
Add a comment...

Nita Vesa

Shared publicly  - 
I just got myself a new phone, the +LG Electronics  G2. I agree with what more-or-less every review says about it, ie. it's plenty fast, its camera handles even poorly-lit in-doors surprisingly well, its battery-life is excellent and so on. I don't personally mind the non-standard placement of the volume-rocker and power-button either, it doesn't take much more than a few minutes to adjust to their location. Basically, the hardware is all good, if not perhaps quite dull. However....

Well, LG screwed up with the Finnish localizations. There are horrible spelling-mistakes all over the place, some sentences aren't even finished at all, and here and there I notice some sentences only partially translated. It shines immediately through that all the Finnish localizations were done by a machine, most likely Google's on Translate or something similar.

There is some voice-operated stuff on the phone, too, but rather predictably it is all only in English. This "Cheese Shutter" in the camera-app lets you take pictures by saying out loud a keyword -- a simple, but useful feature, IMHO -- but it only accepts one of 5 pre-defined keywords, all in English, and none of them can be changed to suit your tastes. Similarly, LG has a voice-assistant-thingie that can send messages, call people, launch some apps and such via voice, but again, it's all in English, and rather stupidly you can't change the command that brings the app up to a custom phrase, you can only choose from "Hello Genie" or "LG Mobile" -- neither of which is fun to say out loud.

What's even more worrisome is the fact that LG has put up an app called "LG Smartworld" on there, but when you launch it and it asks for your country of residence... Finland, Sweden, Norway all are absent from the list. Why even include such an app there in the first place if it doesn't support the country where you sell the damn thing?

Sure, Finland is a small market and in the global scheme of things we mean little, but using a damn machine to translate your texts and not even bothering to double-check them is just disgraceful, it gives a very sloppy image of LG. All about the software screams "half-done." It makes it difficult to trust them and even more difficult to recommend their phones to anyone. I'll grant them that e.g. Samsung's S Voice still doesn't support Finnish as far as I can tell, but other than that their software at least is more-or-less properly translated.
Luc Jansen's profile photoAdam McNutt's profile photoMinna Mäkinen's profile photoNita Vesa's profile photo
+Nita Vesa Misery loves company :)
Add a comment...

Nita Vesa

Shared publicly  - 
Sometimes I can only sigh and bury my face in my hands. (wrt.

To quote the text from Gamespot:
Independent developer Jason Rohrer believes game sales are bad for players and developers alike. In a lengthy and detailed blog post concerning his new game The Castle Doctrine, Rohrer outlined his thinking and explained why he plans to increasingly charge more for his latest game.
"To put it bluntly: sales screw your fans," Rohrer said. "Your fans love your games and eagerly await your next release. They want to get your game as soon as it comes out, at full price. But they are foolish to do that, because a sale is right around the corner."
Rohrer explained that sales can create a scenario where players who just paid full price for a game might feel like they've been taken for a ride. He further argues that because sales have become so commonplace (Valve itself runs five sales annually), the gaming community is now encouraged to hold off on their purchase until a sale pops up.
"This waiting game is likely decimating your player base and critical mass at launch by spreading new players out over time," Rohrer said. "And your fans, who are silly enough to buy the game at launch and waste money, get to participate in a weaker, smaller player community."
In addition, a culture of frequent sales can lead to a no-win situation for developers, he argued. Although developers can choose whether or not to put their game on sale, when so many other creators discount their games, you have to do the same to compete, he said.
Rohrer's own game, The Castle Doctrine, will employ an "ever rising" pricing model inspired by Minecraft, which increased in price throughout its transition from beta to full game. You can buy the game today, in its alpha state, for $8. During launch week on Steam you can get the game for $12, and after that, you'll pay the full $16 for the game.
"The rising price model is really just an inversion of the sales model," Rohrer said. "You get revenue spikes later in the life of the game, right before announced price hikes, which are very similar to the spikes induced by putting a game on sale. But there are no surprises, so no one feels screwed by the process."
"Anyone feel burned by that plan?"


Now, this guy tries to veil his plan of an ever-rising sales model as an attempt to fix the "unfairness" of discounts and sales towards people who bought the game at full price, but he deliberately avoids mentioning the "unfairness" of his plan towards the people who would totally miss the limited-time discounts and sales, either due to not even knowing about the game or not willing to invest in a game in alpha-stage before knowing whether it's worth the investment in the first place. He's just punishing people for not knowing about his game or for trying to save a buck, just as the traditional model "punishes" people who don't wish to wait for sales.

Also, besides the whole "unfairness" - business his plan caters to people who already are devout fans -- what motivation would people who aren't familiar with his stuff or who aren't all-that-devout fans of his have to invest early-on in his stuff? What motivation would they have to invest in his stuff later on, knowing that it'll never be on sale? If it is never on sale the people who are on the brink of trying or not trying it will never go over the threshold, and the people who just aren't willing to pay the full price for his game will simply steer themselves and their wallets towards more appealing, and likely cheaper, alternatives.

The traditional model may hurt those who are impatient and really devout, those who want the game immediately, but on the other hand it rewards those who have some patience, whether they are familiar with the stuff from before or whether they're completely new to it. And well, new players, newcomers to your stuff, are what keeps your userbase growing, not the existing ones.

His plan, really, is geared towards ensuring steady trickle of profits, but it relies on having an already-existing, massive fanbase. It doesn't serve people who aren't his fans, and at least from a personal point of view it is more of a deterrent; at least I have no wish to take part in such folly, I am one of the people who don't like throwing money at things before they're finished or before I know what I'm getting, and with so much competition I'd rather then throw my money at something equally-pleasing, but cheaper, even if it means waiting for a while.
Luc Jansen's profile photoNita Vesa's profile photo
Sure, I knew it wasn't about the price itself, I was just surprised by it.
Add a comment...
Do we have any statisticians here? My friend excitedly linked me to ScareMail ( ) and is apparently all sold up on the idea. To quote the website's description for the application: "ScareMail is a web browser extension that makes email “scary” in order to disrupt NSA surveillance. Extending Google’s Gmail, the work adds to every new email’s signature an algorithmically generated narrative containing a collection of probable NSA search terms. This “story” acts as a trap for NSA programs like PRISM and XKeyscore, forcing them to look at nonsense. Each email’s story is unique in an attempt to avoid automated filtering by NSA search systems."

Now, I tried to explain to my friend all the things I see with this and any similar things that would make it not work as well as he thinks they would:

* If the NSA suddenly sees a spike in hits on e-mails coming from GMail, or from a particular user-agent like e.g. Thunderbird they'll no doubt look into it and notice this kind of a scheme.
* ScareMail adds easily-identifiable text, ie. the mention of "ScareMail" to the e-mail.
* The text generated by it is spelled properly, capitalized properly, punctuation in the right places and so on. That, alone, would be easily identifiable, but if it differs from the text the owner of the account normally produces that makes it even more apparent.
* The text generated contains multiple hits on every single sentence, and all the sentences it generates are all in one, long paragraph. That's statistically a ridiculously large spike in such a small amount of text.
* Such a sudden increase in hits from an account that has never exhibited any symptoms before will also immediately raise a flag about such misleading-attempts.
* If all the replies to the e-mail account are void of hits or they contain a similarly ridiculous amount of hits that'll, again, show up in statistics.
* All these attempts suddenly radically alter the style of conversation when compared to previous conversations.
* The NSA has access to more than just your e-mails, including most peoples' web-browsing history, contacts, metadata from various IM-applications and whatnot, and all these, again, makes it easy to notice any such misleading-attempts in a singular source of data.

I am not a statistician and my friend doesn't listen to me, he insists on something like this being a good, workable solution, and thus I am wondering if there's anyone with actual credentials willing to chime in on this topic?

Hashtags: #statistics #bigdata #email   #NSA  
Minna Mäkinen's profile photoNita Vesa's profile photo
+Minna Mäkinen There's actually quite a lot of really smart, talented people here with credentials from a lot of different sectors. I wouldn't be surprised at all to see statisticians among them.

As for this, well, it's not even about cryptography. Most e-mail is unencrypted and if NSA does have some sort of a back-door access to Google's internal network it doesn't matter even if the e-mails were encrypted in the first place. This is about matching patterns in the e-mail body.
Add a comment...

Nita Vesa

Shared publicly  - 
I just read the news that Google is now filtering child porn - related results rather heavily in English - speaking countries and they're planning to roll out the changes to everyone else soon enough. While protecting children is a worthy goal I believe we are doing it the wrong way as this method doesn't fix the problem in the first place, it doesn't cure the problem, and it will only be abused later on to filter an ever increasing amount of stuff.

IMHO the proper solution to paedophilia would be to re-work how it's treated. I mean, ATM paedophiles are vilified, denied access to multitude of places, often facing abuse even from government officials and whatnot, so such obviously doesn't encourage them to come out and seek help. Instead we should treat the individuals who do come out, who do admit they are paedophiles and register themselves as such officially and maintain a proper connection to therapeutic facilities should be respected for the courage that such an action have taken and we should find ways of encouraging such behaviour.

Banning child pornography isn't going to fix the people who do have paedophilic tendencies, but what if instead we allowed these above-mentioned registered individuals to legally consume fictitious child porn -- comics, cartoons, etc. -- that don't feature actual living people? Even allow these paedophiles to legally exchange this material between other registered ones? These people would hopefully then be able to keep their urges in check and they wouldn't have to fear reprimands if they admitted to their therapists that they're having trouble doing that -- we could then work on trying to find some other, longer-lasting solution for them.

Naming, shaming, abusing, vilifying and even killing them will only drive them ever deeper into the shadows, not out of them, and driving them into shadows will mean they'll also drag any affected children with them in there.
ヽポール's profile photoNita Vesa's profile photoJohn Bump's profile photo
The Prime Minster has demonstrated his complete and utter lack of understanding when it comes to technology and internets. This is just a pathetic attempt to be seen to be doing something about a problem he has no idea how to fix.
Add a comment...
Have her in circles
140 people

Nita Vesa

Shared publicly  - 
I was chatting with my friend the other day about things and he mentioned an application that piqued his interest. It's a simplistic CPU-benchmarking and comparison tool and while it probably does serve its purpose well enough, I couldn't help but criticize how it just gives one, single score for all CPUs. That led me to think about how to create the most-comprehensive,  all-around best CPU-benchmark?

Out of the top of my head I came up with all of the following things:
* It should test every core separately and then all the cores available in all the various combinations so as to see if there's any difference between the speeds between the cores. Also, it would show if there's any interplay between the cores, like e.g. if you're fully-taxing core number 1 does it reduce the speeds of core number 5?
* The benchmarking-code should fit within the processor's cache so as to test the CPU-speed and the speed of its cache, not RAM-speeds.
* There should be several different kinds of payloads to be run, like e.g. floating-point maths, vectors, just your run-of-the-mill arithmetics, plain, old moving of data from one location to another and so on. And again, all these should be run in all the different combinations, so as to find out what each core is best at and if there' is any interplay between these, like e.g. if all the cores share the same FPU then running more than one FP-heavy payload would likely reduce the performance of all the other cores.
* In case some payload was designed to use RAM and not do all of its work inside the cache the score should probably be averaged in part on RAM-timings and clock-speeds so as to be comparable between systems with different kinds of RAM.
* The payloads should also probably come in basic, lowest-common-denominator kind and CPU-specific, optimized versions, like e.g. generic x86-64 payloads and Core2 - optimized payloads. The first kind would be more-or-less comparable directly between different CPUs, but it would most likely not take advantage of all the processor's advanced features and therefore wouldn't fully bring out its top performance. Similarly, the optimized version would only partially reflect the processor's advanced features, and partially the quality of the compiler's optimization routines. In other words, both kinds would be somewhat misleading, and I don't if this problem can ever be truly solved.

What else? What things should be taken into account with the theoretical The Best CPU-benchmark In The World(TM) ? I'm not planning to create one, but I would still like some ideas and discussion on the topic.
ヽポール's profile photoJohn Bump's profile photo
Most CPUs have the RAM controller built in now, so it should be measured. I think what it boils down to is at best you can benchmark an entire system running a certain app, but trying to benchmark a particular CPU in isolation in a generic way is never going to work.
Add a comment...

Nita Vesa

Shared publicly  - 
"Why does Google keep making products for nobody?" --

Why, indeed? Well, first of all, you aren't the customer. You are a data-source. Secondly, GMail, Google+, whatnot -- they aren't Google's products, they're just services for generating the product, and the product itself is...statistical data. Hoards of data, in fact.

The customers for Google's products are the parties buying statistical data from Google or the advertisers using Google's services. Their primary concern is the collection of statistics and selling that out -- how many people visit this or that content at any given time of a day, a month, a year, from what region, what sorts of events coincide with these visits, if there is a growing or declining trend, and so on that other parties can then buy and use for research-purposes, for marketing, for planning future-venues or whatever.

For the end-users of their services, ie. "data-sources, " it's in Google's best interests to make things that are interesting, can be seen as "altruistic" so as to generate goodwill, flashy things that raise Google's name from time to time on peoples' lips -- word-of-mouth is one of Google's most important marketing-machines for their services, and word-of-mouth only works well if the message being delivered in mostly positive. As such all these long-term projects and whatnot make perfect sense, they're catchy, they're trendy, they make Google look good in the eyes of the general public, and that correlates with more users for their services and therefore more data. Even if what the author here complains about, ie. these contact-lenses and whatnot, aren't exactly a silver-bullet to this or that problem they serve as PR for Google and, like in any research, you need to take steps that don't produce anything useable in order to get to the later steps that do -- the author completely ignores this and just wants instant gratification.

So, this leads us back to the question: why does Google keep making products for nobody? Or, why do you expect Google to singularly cater to your needs and wishes, to only focus on short-term instant gratification and ignore any long-term business plans, research that may yield them lucrative patent-deals or better services?
ヽポール's profile photoNita Vesa's profile photoRJFerret Randy's profile photo
+ヽポール "Clearly we are their customers because without us there would be no ad revenue." -- That doesn't make us their customers. In the case of ads their customers are the ones paying for the ads and the statistics behind them, we are the consumers.
Add a comment...

Nita Vesa

Shared publicly  - 
I stumbled upon Facerig ( ) today. It's basically software that monitors your facial features and head position and mimics those on virtual characters in real-time. It's not the first one to try this and certainly not the last one, but who knows, maybe this time it'll stick?

The obvious place where this stuff could be used would be in any game offering multiplayer, be it just 2 players or a full-blown 25k players MMO. It sure would add some interactivity and depth if your characters reflected your own reactions, even if it didn't bring anything actually useful to the table. Imagine e.g. TERA, or World of Warcraft, or, say, Guild Wars 2 picking this stuff up and reflecting your facial expressions in real-time inside the game to anyone who happens to be looking at you? Personally I would find that very endearing.

The other place that immediately sprung up to my mind where this stuff will almost certainly pick up speed like there was no tomorrow is... all the various otaku-scenes. I mean, these guys and gals are willing to even marry virtual characters for real, so you can bet they'd be willing to part with their money in order to be able to pose as cutesy anime girls on the various chats. And do other things... cough

Anyways, it looks cool, and even if I'm not exactly excited about this I feel it'd be a good addition to our current games and therefore I hope to see it gaining in popularity.
ヽポール's profile photoNita Vesa's profile photo
It was half sarcastic, I know what you are like. Harassed by pervs all the

I don't think it's necessarily true about men having more sex drive though.
If you hang around in mmorpgs with teenage boys or read to much 4chan that
is the impression you get though. Even outside of those venues advertisers
love sex because it's easy, so it is used everywhere and it seems like men
think of little else. From that you could conclude that all women are
extremely vein and primarily interested in beauty products and Photoshop.

Also, frankly, what do you know about men? ;-)

I take your point about it not being a negative thing though.
Add a comment...

Nita Vesa

Shared publicly  - 
Reading up on +Valve 's Steam Box and the Ibuypower's take on it prompted me to try and benchmark Metro: Last Light on Windows and Linux and compare the results. While waiting for Steam to finish downloading and installing the game I've been trying to get surround sound working under Linux.

Now, let me say that Linux fucking sucks on the desktop. I can't find any way of making 5.1 surround sound over SDPIF working -- I'm not talking about pass-through -- even after trying to follow the instructions at . If I do as the instructions say Pulseaudio crashes and does not recover. And how's a Joe Blow supposed to get surround sound working if even a geek can't? Also, it seems Pavucontrol is the go-to mixer application these days, but it seemingly lists all the cards in random order, it's confusing, messy and I can't seem to be able to designate which card in my system I want to use; there's a button that says "Set as secondary," but it doesn't do anything, only the button itself changes between pressed and de-pressed, and resets itself anyways whenever I restart Pavucontrol.

Audio output has been a thing on PCs for several decades now, and we still can't get it right under Linux or actually coherent, user-friendly mixers, even?
Luc Jansen's profile photoNita Vesa's profile photo
You'd make the news if it did :)
Add a comment...

Nita Vesa

Shared publicly  - 
Recently NVIDIA introduced a new feature in their Geforce Experience - application called 'Shadowplay.' Shadowplay is a feature that lets you record video in-game and use the graphics hardware for the actual encoding of it, resulting in lighter experience than with all the software-based recording applications. I just tried it myself earlier today and by golly, it sure does what it says on the can: it records 1080p@60FPS video and I didn't notice any drop in framerates at all while it was recording. None, whatsoever. Quite nice. Also, the video quality itself seems more than reasonable, it's actually pretty good.

There are a few downsides to it still, but as it's still in beta I would suppose NVIDIA will work on that; it only records audio in stereo, no surround-sound, and it's limited to 4GB videos. is a link to a sample I recorded, though playing it in VLC on my PC results in choppy playback. MPC HC and WMP play it smooth.

If you own a recent NVIDIA GPU and like to record stuff this might actually be worth checking out.
Nita Vesa's profile photo
I've been playing some more with the Shadowplay - thing and it sure seems to work well, and even the lowest quality - setting is still perfectly reasonable, it's still watchable as-is. This Shadowplay could turn out to be a really great thing for gamers on +NVIDIA  - hardware if only the company would get off their asses and realize the possibilities here: add support for streaming to Twitch or something so that you can have high-quality live streaming that doesn't bog down your system, add support for sending to YouTube straight from the app, find a service where you can upload the full 60FPS - videos and play them back at that speed instead of lowering framerates -- this would be great for showing off special effects or really hectic battles and such.

Also, this makes me wonder if 3D stereoscopic recording would also be a possibility? I know, I know, people keep saying it's a gimmick, it's not worth the effort and whatnot, but there are us suckers here and there who like the stuff and would like to both produce and consume such content. +NVIDIA could try to push for one or another service to start offering 3D - content if they wanted to, and games are one of the easiest sources to obtain such content from..
Add a comment...
Have her in circles
140 people
Map of the places this user has livedMap of the places this user has livedMap of the places this user has lived
Pori - Orimattila - Orimattila - Lahti - Pori - Tennevoll - Heinola
Haters gonna hate.
A pansexual gamer chick who loves RPG games the most, though doesn't shy away from occasional slaughter of innocents or helping the walking dead become permanently non-walking ones.

General interests include obviously gaming and as should be evident from my posts it also includes all kinds of topic related to computers and computing. Also medical research, copyrights and basic human rights, morals and ethics, angry EBM music, and movies; I have likely seen closer to a thousand movies in my life so far and I'm still not satisfied. Oh, and female breasts are included too, you mustn't forget those. Male breasts aren't nearly as enticing.
Bragging rights
None. If I die some day as a wrinkled, cynical bitch with a smile on my face then I view that I have accomplished something worth mentioning.
  • Erkko-lukio
  • Orimattilan yläaste
Basic Information
Looking for
Friends, Networking