treffer

For a standardized language like SQL you don't have API incompatibility.

Yet a cost based optimiser may break your application, completely.

Calendar Versioning prompts to update and highlights that you should test the behaviour every time, which sounds about right.

Also even a patch release of a SemVer library is allowed to be runtime incompatible. Throwing an exception on incorrect usage is down to the patch definition. It fixes incorrect (undefined) behaviour. Going from a O(1) to. O(n²) implementation to fix behaviour is also within SemVer patch level.

SemVer does not guarantee that your code will continue to work as intended after an upgrade. It only guarantees you don't have to change your code between minor versions.

Actually sorting should be fine. The order of UTF-8 will always be the same as the codepoints encoded under it if you sort it as binary.

I really appreciate this property of UTF-8, and I can highly recommend to do a pen&paper exercise to see why it preserves order :-)

It wouldn't work if you want any type of normalisation and especially solving composition/decomposition.

Upgraded all my devices at home, worked flawlessly.

Any news about 19.01?

There were several articles around that...

It was a very special manufacturing procedure, something that is hard to copy without the special hardware.

And Apple bought all that supply. You couldn't get the production line to do it...

(IIRC, bit my memory may fail me on that. Please Google it, I would be interested in rereading it!)

In the EU you can use https://geizhals.eu/?cat=nb which is pretty decent. I'm using that regularly to check options.

Is something like this not available in the US/other territories? If not: here is your opportunity :-D

My must-have list is https everywhere (from EFF) and uBlock origin.

I hope that the mobile browser will get at least the same level of attention as in the past.

Start by using done service discovery. That way service names stay the same but service implementations and locations can move.

Then ship the static list of names (should be short) and per-service credentials (highly highly recommended).

Another pattern is co-locating a proxy with your app. See e.g. linkerd on how to do that. This will also unify the handling of circuit breakers and connection pools across services - even without any shared code!

I have a somehow different view on swap.

The issue is not swap or swap utilisation, the problem is worst case latency. Even for a database an OOM is usually better than a latency hit that makes it unusable slow.

As a simple example an app might start allocating and use memory in an infinite loop. How long will that take? How long will your system be unresponsive?

If you have more swap than you can write in 30s you'll most likely do it wrong (your system can be unresponsive for 60+s).

Another worst case would be allocating all memory, using it and then performing random reads throughout the memory space. Your swap to ram ratio defines how much misses and thus how much IO you are doing instead of direct memory access. This should stay way below your IO capacity.

As a result I usually try to use a small swap partition and monitor for swap-ins, not swap usage or swap out.

So that's my thinking around swap mainly due to the fact that I have seen too many servers causing issues due to swap related latency.

1. There is already a company doing that, among other things: CoreOS

2. There is usually not a "natural" conflict for companies that started out of open source projects

3. A full kubernetes deploy is already hard

I'm actually quite happy that we will now see Heptio and CoreOS pushing for boxed kubernetes deploys. Multiple entities pushing an open source core is usually a good thing.

Well, errors at a rate of one every few terabyte, and most of the time it was mainly decompression, and always caused program abort iirc.

But those have been a few years ago, so my best advice would be: retest.

You are testing your archives anyway, right? ;-)

pbzip2 only does parallel compression. lbzip2 can parallelize compression and decompression.

With both pbzip2 and lbzip2 I got errors every now and then while everything worked with bzip2. YMMV.

Also note that bzip2 is block based (due to bwt) and thus does not compromise compression ratios like parallelized gzip implantations.

At 32 cores you can saturate Gbit links (even with good compression ratios!). I hope we will see some more bzip2 love in the future due to it's perfect fit for parallelism.

You can ping Barton (the project Sputnik head) on twitter about pricing and availability bugs. He usually gets them sorted quickly and is happy about feedback. I've reported issues when the XPS13 launched. It works. They fix this stuff.

But keep in mind that the EU/Prague pricing might be different once launched there.

I have a QHD version. I have Firefox set to 125% (or 150%). I'm heavily relying on <ctrl> +/- to control the information density in terminals and browsers. That works really well.

Most UIs nowerdays adept to DPI, so no real problems there (idea/intellij just fixed their stuff to work better on HiDPI).

It's today just a matter of choice. I usually use a prescaling/font size etc. that does not pack too much information onto the screen but I enjoy that I can increase the density if needed!

The touchpad works well with multi-tap and can handle single/double/tripple tab. Feel free to map that to the middle mouse if needed.

The keyboard is actually quite good once you get used to it. I used to need a keyboard replacement every 12-18 months on my lenovo (yes, you read that correctly). The Dell one has now survived 12 months, so I'd say the build quality is quite good, perhaps even better than the lenovo one (might also be due to design - nothing gets stuck under the keys, which helps a lot).

Dell has even send me replacement parts in the past to repair stuff on my own.

[EDIT] type.

The only cards that support DFS are atheros ath10k (I've tried DFS frequencies with an openwrt AP).

Anyway, I've yet to see an ath10k mini-pcie cards that would fit into this machine. The wireless can be easily replaced (at least on last years machine), so hardware hints welcome.

XPS13 / QHD / Touch user here (last years model). I've used it rarely but it works without flaws. (Well, firefox can't really handle multitouch on X11, but that's a whole different story).

Only problem is that it's incompatible with kids. "Hey, what's that? TAP"

Oh, let me press reply on the screen :-)

There is an online cancellation service in germany that will help you to cancel just about every contract: https://www.aboalarm.de/

They have ready-made templates for many compaines (esp. ensurance and telco ones) where you just fill in your details, sign it via your touchpad and let them send it in a conforming way.

I /think/ the business model is focused on post-cancellation: they'll show you other options for the canceled contract.

I'm stunned that this does not exist for the US. It seems to work in Germany.

Why would you want to go to a gas station?

Batteries improve ~15% per year ($$$ / kWh stored). It appears to be some sort of moore's law for batteries. It also means we'll see way better batteries in 10 years (4x better regarding $$$/kWh).

Now with way more range and non-stop travel (2-4x) the bottleneck will be how long you're willing to sit in your car without a major break (sleeping).

Such a future would be very similar to charging your phone at night. Most people can get away with just charging at night. Even if they travel. I expect the car battery capacity to grow beyond the point where intraday charging is relevant at all.

And once we've reached that future going to a gas station will seem like a strange thing of the past. Yep, it might be 10-20 years from now. But we'll hopefully all watch this happen.

Which has it's own set of problems, btw. Try to find an API for "delete".

Oh, and I'm eager to see how their API survives mutual parties extending it. (There is a reason why namespaces exists)

Namespaces is something I consider a strong plus for XMPP. You may want to extend the data model at some point. You need a way to say "this is my extension, please ignore it". Or at least you need it if you want multiple implementations to work together.

This makes it possible to experiment over the live xmpp network.

map/filter/reduce etc is possible with XML. Scala does it. Other languages can do it to. It's purely depending on your language. Even JavaScript has better XML support than the language you're portraying here :-)

Doesn't this contradict an interview with Senior Vice President of People Operations

http://www.wired.com/2015/04/hire-like-google/

Preallocating memory is usually an opimization for throughput.

Servers can have up to 1TB of ram without becoming overpriced.

But starving 80k users due to low buffering will be expensive in the long run. Way more expensive than RAM ;-)

I'm from germany and spend some time at the technical university of munich. This statement was not accurate at the time of writing and is even less accurate today.

There is a well known university ranking in germany, named "Hochschulranking". An older version for sports can be seen at e.g. https://www.sport.uni-freiburg.de/bachelor-ranking2012.pdf There was no such thing as equal quality, obviously, just a baseline of what you should learn. This baseline is from what I've heard higher than in the US.

Nowerdays universities are praised for their specializations and receive extra money for those activities. See https://en.wikipedia.org/wiki/German_Universities_Excellence...

German universities do pop up in global rankings, too. https://www.timeshighereducation.co.uk/world-university-rank... lists the technical university of munich as #28. (How biased those lists are is another topic)

One more thing, I'm currently trying to find the source again, but ~20% of foreign students decide to stay in Germany after receiving free education. Those numbers were way lower in the US iirc. Given such a high number of staying students I'd say it's well worth the money (for Germany).

The classic example is the reverse engineering of the IBM BIOS for 'IBM compatible' machines. And it was done with 2 teams, one analyzing and one implementing.

But that predates my existence :-)

HTTP: here is the length (response header) and here is the payload transfered with checksums (tcp). HTTP has quite good integrity of the data you requested.

If you mean HTTPs is great to check if it's coming from the right server (identity/authority) then well, see how well the CA system works and how weak certificates really are (e.g. "Ron was wrong, Whit is right"). You can work around that by doing certificate pinning (see google chrome for example) and stuff like that, but in that case I'd already be better off signing the payload and sending that as part of the response.

The scientific institutes of germany worked around all the CA headaches by becoming an intermediate CA (see who signed https://www.pki.dfn.de/ ). I'd just expect more intermediate CAs popping up as a result of HTTPs only which will weaken HTTPs even more.

(And yes, I'm using https everywhere and would enable https for just about every site, except e.g. for linux iso downloads and alike)

Huh, I have everything on github if you are interesting. https://github.com/gfreed

I am still using the blobs, but I was thinking about gradually replacing parts once replicant reaches a good-enough (e.g. samsung rild, have to try it though).

Btw, if you are interested in building the ROM add me on jabber (arty@jabber.ccc.de). I want to build the next version this weekend anyway and I may as well guide you through building roms, too. Plus I would appreciate help on porting to new devices :-P

[EDIT] I'm unhappy with the ROM scene, too. That's why I started my own ROM in the first place.

Not that easy. I'm trying that right now on a second phone. I've rolled my own ROM which I'll update this weekend with all the cool new stuff (especially the mock play services / maps v2): http://forum.xda-developers.com/galaxy-nexus/development/rom...

mar-v-in and his µg are corner stones to build a nice free android. As for a server side I'd suggest ownCloud at the moment. I'm running one on ipv6 at home. Works great.

1. My ISP send me a router that has a feature to host a internet accessible fileserver. Out-of-the-box. ISPs in germany do not promote home-hosting but they give you devices that do it.

2. Incoming / Outgoing TCP/25 blocking: Not a problem with ISPs in Germany.

3. Blacklisting: At most a problem if you send mails. Plus GMAIL requires SPF/DKIM for just about evey IP, so yes, you are on a blacklist unless you do some DNS magic. BUT once you do the magic it will override IP block - unless the other side has a shitty setup....

Buy a domain on a service that allows API-based DNS updates.

Put up a RPI at home. postfix + dovecot + roundcube should do the trick.

Add FW forwardings for 80/443/25 (or allow IPv6 to pass through)

Update DNS records every N minutes. (cron, nsupdate, dyndns clients, amazon command line tools....). You will need SPF/DKIM.

The RFC for SMTP says Mail Servers have to retry for 7 days before giving up on mail delivery. This should be plenty for your home server. There are also commercial Mail relay and backup MX services (sometimes even as a free offer for buying domains on website X).

You can backup the SD-Card whenever you want. Your Mail stays at locations you control.

I currently have a root server, but I'm heavily considering "in-housing" those services because of the NSA activities.

EDIT: it's 7 days PS: Some old firewalls block dynamic IPs for mail delivery. I'm not sure how common this is today, especially as SPAM and botnets have evolved a lot.

We were thinking about it. I fired up our apps testsuite and found a bug: https://mariadb.atlassian.net/browse/MDEV-4711

The bug was now been analyzed but not fixed for ~3 releases.

That said I also ran the testsuite against 5.6 and found some problems, too. You might also be concerned about http://bugs.mysql.com/bug.php?id=69274 - this means that if you loose an idb file during a crash you're out of luck to use your db ever again.

We're thus currently stuck at percona server 5.5 with regular investigation of the alternatives.