justin_hancock

The expectation that a machine should not make these mistakes is reasoned, humans might make these mistakes but they will generally realise their mistake. The machine doesn't know its wrong. A human learns to drive within 20-40 hours with reasonable competency, versus the effectively millions of hours of computation that's used to train the models, Given this the expectations aren't as unrealistic as some on here claim.

Disagree on latency, for certain classes of applications it really matters. Latency can be caused by a myriad of things, but if that thing is your means for cut over, you haven't failed to respond, you've failed to respond in a timely way. The telco space has strict latency requirements. If you have a failure which results in erratic behaviour on the network due to the database taking time to handle a node failure you aren't in a good place, it has a financial impact.

There are databases that can provide high consistency and availability.

Volt is another good example of solving this well. It's consistency guarantees are strong, and offers very high availability through it's replication which commits transactions to replicas simultaneously. In theory it's weaker on the partition side, but in practise with modern hardware and networks it's impact is not felt. It also offers multi data centre replication, with all sites being active.

Bournemouth has a long runway, legacy of its WWII roots. It was chartered for £60,000 for some 20-minute subsonic flights. There were also some day trips to Cairo from Bournemouth.

It is a real issue, I was amazed to find places such as Milton Keynes were seeing foreign investment on new build - friend is renting there and his new build place is owned by a family based in the UAE

Zwift and Bkool already have offerings in this area. Zwift is outstanding.

It is brutal and in some countries, likely to result in a law suit as it can be construed as 'Constructive Dismissal'. Extraordinarily damaging to the person on the receiving end. Its better to do it the right way, make the position redundant, and enable everyone to move on.

I think you've hit the nail on the head with Scala. It feels every conceivable concept in language research has been thrown in. The lack of coherency is sometimes quite jarring.

Coherency/conceptual integrity are things I look for in computing environments they make reasoning things much simpler as logical deductions can be made, I don't find myself doing that with Scala.

Warrants are required in the UK, though there are special exceptions such as when a serious or dangerous incident, agreed its open ended but the onus is on the Police officer to prove it and entering without one is unheard of in my experience, the source below provides more detail, and is a reliable source:

http://www.adviceguide.org.uk/england/law_e/law_legal_system...

They're useful where you need to grow a collection in constant time. Re-sizing arrays is potentially very expensive.

The one thing seems to have been omitted is housing costs? I am in the UK and housing costs have risen very sharply thus eliminating any other reductions on pricing. This is one area the current generation is much worse off, rents are high, and purchase prices even higher. My parents with a single relatively low income could afford a house, my sibling and his partner with two reasonable incomes can't afford a one bedroom apartment.

Some parts of the UK do have cheaper housing but with good reason, low levels of employment.

Interested to hear the US and wider perspectives.

HBase 90.4, problems with I/O we had a very heavy read load on top of a write load, write load bursting 14,000 TX per second, and an average of 8,000 per second - each record around 2k.

Because of the I/O the WAL had to be turned off, this introduced problems when Region Servers occasionally died. Implementation of large regions 10GB, and fairly large HBlocks 512MB, increasing flush sizes to reduce minor compactions. Use of MSLAB to virtually eliminate GC all together, use of large heap 12GB on RS.

Worst problems we experienced was META corruption, that really , really sucked.

Also they used an old version of HBase, 90.4 in the benchmark, HBase 94.0 has been out since May 2012, and has a lot of work to improve performance especially remove lock contention on the region server.

My understanding of facebooks Pod Architecture for HBase was not the name node but simply scaling HBase, HBase gets rather unpleasant at facebook scales. The facebook HBase fork has things like compactions disabled to improve performance.

I ran a HBase cluster with 1PB storage, it became very unwieldy at this scale, thousands of regions and lots of tricks to keep it happy. As for SPOF, the name node now has HA and it works very well.

Bruce Eckel is releasing a book this year on Scala, Atomic Scala: http://www.atomicscala.com/

Eckel's Thinking in Java was a classic, I suspect he will this one out of the bag too.

The Odersky book is good as a reference, but not a great teaching tool, its too big. I watched the video lectures, they're quite good if a little dry and occasionally quite abstract especially if you're trying to apply Scala to real world problems.

Be warned though interop isn't perfect between Java and Scala, sometimes you have to resort to tricks to make things work, Hadoop APIs being a good example.

I find this quite curious, how would it work with clothing? I mean most clothes designs are the work of high end fashion houses, the middle market and lower end then produce very similar styles, its an accepted system. Surely this mean the fashion industry would be subject to the same design patents and thus the middle market could be legitimately sued?

This is a just discussion of the hype cycle, you can apply the very general rules to almost any technology. No new insight here, it feels like link bait.

I have had the joy of interviewing people who've lied on their CVs (Resumes). One particular pearl was a guy claiming he'd worked on a particular project I had lead, I proceeded to ask him about the project who ran it, what it was about etc, he gave some vague answers, needless to say he didn't get the job.

Though the worst offenders for this aren't individual agents, its usually 3rd parties such as body shops trying to provide augmentation staff or recruitment agents doctoring resumes to try and get candidates in. I believe it was the case with my earlier example.

Although its not on mobile didn't Google's desktop search tools cover an awful lot of this? I certainly remember it returning, emails, documents and other textual data in the result. It also had plugin support.

It may not have existed on mobile but one could argue that much of it did already exist on the desktop, bringing similar functionality to a mobile is not especially innovative in my opinion, its a feature at best. Integrating things is not innovative either.

I would love to see the fireworks of Apple taking Google to court on search patent infringement.

Google as far as I based android on the Apache VM which was implemented using the published OpCode spec for Java. So for ripping off the OpCode/VM, actually they did something quite legitimate. From memory the Oracle claim centers on a few specific classes that they claim Google reverse engineered from the original Java source.

As for the incompatibility it was Oracle that chose to deny Apache a license for the Harmony VM, this was because they knew it was a threat to J2ME.

The article is very long, I do apologise for that. However its a very thorough analysis of the economic issues of our times.

The other critical factor is whether the acquisition is a cash deal or a share based deal, a share arrangement would be very efficient indeed, hard to find meaningful details of the acquisition.

The problem is knowing which part of Motorola held the patents, if it was mobility, which would seem logical in some ways (encryption of video streams and media) it would be a boon. Microsoft have a big case against Motorola over the license fees for h264.

Plus I assume there's the iDEN patents, which I suspect will have lots in the download/media area. It will certainly keep the wolves from the door.

My big hope is that Google do something altruistic with this, however it's shareholders may have other ideas.

Does this include any of the H264 patents that Motorola hold? If so this makes this story even more interesting, the cases for WebM/VP8 would change substantially.

I think that some people have missed some of the points in the article in particular the aggressive litigation against Samsung, HTC, Motorola etc it isn't just Google thats being targeted. This is an anti-competitive stance, Apple, Oracle and Microsoft are being very aggressive in their pursuits. In my opinion the critical and genuine patents in mobile are linked to the radio portion of the device, the remaining patents are dubious as they don't represent novel or non-obvious solutions to a problem, more I did it this particular way so I am patenting it.

The US patent system, is broken. The cases as far as I know are being fought in the US, I doubt any of the claims would be accepted by a European court, though feel free to correct me.

In my opinion just some paraphrasing of the old Sun documentation from several years ago. You're better to read the source material to fully understand what's happening in the VM, its a bit incomprehensible at first but a couple of reads and it really sinks in.

I find it all rather sad. BP has committed an appalling act of negligence however the parties operating the rig and the company that provided the Blow Out Preventer (Trans Ocean and Halliburton) still seem to be avoiding the worst of the flack. They are culpable too, yet the administration chooses to ignore it. This has been a marvellous political tool for them.

Wonder if Dow Chemical has fixed the issues in Bhopal yet?

My PC I had for University, bought in late 1994, had 16MB RAM (which was huge), typical PC in that era ('92-94) was 4MB RAM, which was considered the minimum to run Window 3.1. I think the 8MB figures are a little wide of the marque, by 1997 most machines I saw were already hitting 32MB upwards, by 1998 many were hitting 128MB at the upper end to run Windows NT 4.0.

Perhaps a means for making an invisibility cloak, e.g. display the image of your environment on the cloak.

Whilst there maybe some nasty sensationalism going on here, that does not entirely exonerate Apple/Nokia. They put enormous pressures on their suppliers to deliver more goods in less time. The inevitable consequence is that the supplier pressures the production line to produce more and nasty shocks like this okay. The fact that this went on for a year or so unchecked means Apple is not auditing regularly enough. If Apple were running the production process itself I doubt this would happen, however when it outsources it does. Apple needs to be a good corporate citizen and also put its money where its mouth is, if suppliers do this cut them off.