1,075 karma · joined April 1, 2013
[ my public key: https://keybase.io/gbl08ma; my proof: https://keybase.io/gbl08ma/sigs/IptL6B3YTB0jyItR6dM36ubj_vlkj4yiJ0ohX-QY0CE ]
It suffers from some graphic bugs (mostly in the properties dialog, for objects in PowerPoint for example; usually you can fix it by reloading the dialog), also, TrueType doesn't look perfect. I can't remember having it crash heavily.
I needed MS Office because while Open/LibreOffice is good enough for most things, and is actually more than sufficient if you write your materials from beginning to end in them and never export to proprietary formats (i.e., always keep saving in ODF), the compatibility layer with MS Office is not perfect and is not sufficient for more advanced things (such as the fancy 3D-and-the-like styling options my colleagues insist on using). The imperfections* of the document compatibility layer become a problem when working on the same document with people who use MS Office.
Using a virtual machine would probably work, too... if I had an infinite amount of Windows licenses. I can't use the license from my real-hardware install, as I still need Windows-on-real-hardware for a few things (to do with drivers and device flashing).
Using Office was the main reason why I had to reboot into Windows, and so I couldn't be happier with this setup. My productivity has increased a lot (no need to wait for reboots). Besides Office, I also use Wine to run the emulator of a platform for which I develop software (the emulator is not open source, and are not available for any Unix but OS X, which I don't have access too). So Wine does a really great job for me, doing things that I think can be called "effective usage" (in fact, I don't run any games under Wine).
* MS Office import/export is actually quite good for a FOSS piece of software, and while the "imperfections" are being perfected as new versions are released, Microsoft too keeps releasing new Office versions with more shiny features, so I understand the difficulty on keeping up with the compatibility on closed formats.
Since this is a public page, with the title "Which Windows operating system am I running?", I think the right thing to show when the user is not running Windows is, well, "You are not running Windows". That would clearly inform users, who may not be aware, of the fact that their OS is not Windows (and that, most likely, that website will have nothing to help them with using their current OS). Another alternative would be "Can't determine Windows version". Anyway, they way things are now, that page is definitely wrong.
Had this been some years ago, and I would say the page was relying on some ActiveX control to check the OS version, and that on error, "Windows 8.1" would be shown. Fortunately, even Microsoft seems to have abandoned ActiveX.
My Linux setup is now very offended because a web page called it names.
We certainly have the technology to make better cards with more storage (BTW, the space for SMS text storage on the cards I own is ridiculously small, too). But as people moved on to store information on "rich" contact systems and, more recently, in the cloud, I think there isn't much motivation to innovate in that area.
I did not cheat, and many of the names could be real (I swear I saw some that were actually titles of real articles, but in fact were bot-generated) - I'm not sure what makes me tell a real title out of a fake one in these situations. On other situations though, it's very easy (grammar makes no sense, there are odd amounts of quotes, etc.), but in my experience these tend to be a minority.
Actually, I find it funny that in my experience, most titles that could be of real, interesting hacks fitting well in the spirit of HN, are the bot generated ones, and the most political/controversial titles (which often you'll see discussed on many places other than HN) are the real ones.
Edit: I wish there was an easy way to record your session, so that later you can analyze each title more carefully and do some math on which options you "upvoted".
Assuming everything worked as described, people could always take video screenshots, or record the screen with a camera (30+ fps is very common these days).
I didn't say that was correct, I was just saying that's what the article (incorrectly, in my opinion) appears to say. I'm not sure who's misunderstanding who...
> If sha256 is broken so badly that arbitrary outputs can be chosen
It's not about sha256 being broken or not, but about the way it is being used for mining (being used incorrectly != broken hashing algo). Even though I agree, if it was being used improperly we would have noticed by now - I could have chosen a better example.
Altcoins, among other things, provide a way for testing (sometimes crazy) ideas that can't be tested on Bitcoin at this stage, and allow for more choices to be given to the user (in the same way Linux is not the only open source kernel, GNU GPL is not the only open source license, etc.).
If you think of Bitcoin as just another open source project, and consider "altcoins" to be forks, you'll find forks that are basically just copies of the main project with some values adjusted to the will of the creator, and you'll find more elaborate forks that actually add something new (even if these additions prove to be inadequate to merge into the main project at some point). The second type of forks is, in my opinion, very desirable, while the first one is pretty much useless and, sometimes, is nothing more than a pump-and-dump scheme.
If the benchmark isn't just for the show off and is actually benchmarking something, then the results are impressive - on my phone (quad-core ARMv7, MTK 6589 at 1.2 GHz) it reports a speedup which is, on average, 7x. Obviously, the benchmark could be just testing some edge case that, when Dalvik gets in the way, goes pretty slowly, when compared to native binaries. Seeing as Google bought FlexyCore, it's probably because it's something serious. Too bad the benefits will probably only be available to future versions of Android to which I won't be able to update my unpopular phone or my damn slow ARMv6 tablet.
It would be interesting to see the page title along each URL, so that one could more easily judge about the content behind each.
According to that API intro page, their databases contain over 1.8 million entries for books alone, so the download would be rather big I think.
Uptime Robot also has an API like Pingdom, but with the added benefit you can monitor up to 50 websites for free. The API allows me to almost never log in to the control panel, as I watch the status using my custom monitoring panels.
The silly bit is the fact that the file extensions are hidden by default, and users can only use the icon to check the file type.
Perhaps the submission link should have been of the far more insightful website http://istruecryptauditedyet.com/ ?
"That’s holding onto the past; and providing a lack of recognition for young people, who will literally have no idea what one is."
I don't see what's the problem with "holding onto the past": everyone knows there was a past, that the world didn't start when they were born. One way or another, we get to know about things that existed (and disappeared) before we were born, and we accept it as something natural.
Going by the line of thought the author suggests, we may as well, for example, ditch files and folders - not just the icons, but the complete organizational concept. I guess we could just use "libraries" and "galleries", but look at where those are going... in some years, using these concepts could as well mean "holding onto the past".
"Imagine a native digital medium"
In my humble opinion, "digital" and "native" don't combine very well. "Digital" is far from being something every human is familiar with, at birth (so that it can be considered "native"). Killing the icon is actually going to have a negative effect on the creation of a "native digital medium", because icons are many times representations of real life, even if poor representations. For the medium to be native, it would need to actually be closer to what we see in real life, than to be even more abstract.
And let's not forget the fact that after knowing their meaning (regardless of them having any meaning in the physical environment around us), icons are an easy and quick way of recognizing certain functionality even on never-before-seen UI screens, which gives a better first time experience of any software.
http://google.com/gwt/x?u=http%3A%2F%2Fwww.rollingstone.com%...
I understand the "type the correct string and go to the next step" kind of tutorial, but as someone who is used to the Linux shell and had heard about Docker, but never used it, it's a bit frustrating to finish the tutorial and not be able to play around a bit with the "simulator" to test Docker without having to install it.
Anyway, I suppose it serves the purpose of "getting started".
I don't read French or Japanese but even the shell outputs are interesting reads.
Not entirely related but I also have worked with WiFi equipment (including that phone) that allows for changing the MAC address, which is something more common I think.