d4nt

In theory a separate company who's core competency is running IT systems should be less likely to mess this up, not more so.

However, retail banks are rapidly becoming sort of API hosting companies for the processing of financial transactions. Their core competency IS running IT systems.

If none of RBS's physical branches opened tomorrow, the UK would probably tick along just fine, with some inconvenience caused to people who deal mostly in cash. But fail to process payments for 24 hours, and the whole UK has problems. If this had gone any longer than it did, it's not hard to envisage a run on the bank and a domino effect to other banks. It's really quite scary how badly we depend on these IT systems, waaay more than Google or AWS.

I work with some people who used to work for RBS IT, they tell me that the department was as bureaucratic as they come (presumably to reduce risk) but once all the hoops had been jumped through, it came down to single team who would put code into production, and they were often quite poor at it.

When I think about the scale of this mess, and the hundreds of millions that RBS is going to loose in compensation and lost customers, I hope that RBS (and in house corporate IT in general) starts to appreciate the need for automated deployment and fault tolerant IT systems rather than yet more process as the answer (I'm looking at you ITIL).

I think the apple was only ever valuable in the sense that it had some calories and could keep you alive a bit longer, long enough to keep on working a bit more and generating other forms of value for others.

Value is a very strange concept, if there were no people then apples would be of no value. If there were people but they all agreed they didn't want apples and weren't prepared to work for them then they wouldn't have any value. They value of apples is defined by how many other people want them and how hard they are prepared to work in return for them. When you eat an apple you stay alive and thereby increase the value of all future apples.

OPM? Other Peoples' Money?

I think it rarely makes sense to view Microsoft as one consistent entity; you have to look at which division is doing this and what are they're motives[1].

The Azure bit may have some affinity to the Windows Server division but probably doesn't feel pressure to sell desktop OS or Microsoft Office licenses. I think Azure wants to attract developers away from GAE and AWS, in that context this move makes perfect sense.

This is no more unusual than MS Office being available on Mac or Silverlight support being somewhat absent from Windows 8. Both things make some degree of sense when you 'follow the money'.

[1]: http://www.businessinsider.com/big-tech-org-charts-2011-6

Good point, when I was looking into Azure about 18 mths ago, you could have a Windows VM but the idea was that it was instantiated and port 80 was called and then it might get thrown away again afterwards. The same may be true for Linux instances, but its an important distinction between this service and AWS/Linode where you can have an instance running long term which has it's own state.

Along similar lines, I wondered whether making up votes have a steadily decreasing effect (say, first up vote is 5 points, then 2.5, then 1.25) and down votes have a steadily increasing effect (0.375, 0.75, 1.25...) would be wise.

Exactly, I recently had a startup with approx 0 users try to recruit me by arguing that building the web app was "programming 101... we need someone who can architect for massive data volumes". I didn't take the job. That kind of attitude is going to cause them serious issues.

    > All human languages have very similar every day vocabularies

I spent a lot of time working in Sweden and had many conversations with the locals (all of whom know English very well). Many of them remarked that English has more alternative words for something than Swedish and that each one has a subtly different meaning. For example, how you can use "grand" in place of "big" or "large", they all mean roughly the same thing but "grand" brings with it additional meaning. So, my personal experience appears to contradict this statement.

My point is really that languages and alphabets are tools and some tools make certain jobs easier. You could write a website in assembly, but its easier in Rails. However, I suspect our ancestors will look back at Rails and have similar feelings to us when we look back at ancient writing systems now.

I acknowledge you know a lot more about Hieroglyphics than me. But my comments were based on the understanding that it was really the development of an alphabet by the Phoenicians that took writing from an obscure system to something usable by every day people. Maybe this isn't the case?

Also English is widely acknowledged to be a very expressive language compared to, say, Swedish which has a sparser every day vocabulary. I'd be surprised if you told me that the nuance of the finer English texts could be adequately captured in Hieroglyphics.

I agree, Excel is a programming environment. Could do with better version control and debugging though.

I think the whole everyone should learn to code this is right, but a little ahead of its time. A little like saying that peasant farmers in the bronze age should learn to read and write. While 100% literacy is now acknowledge as a good thing it took a while to get here. A farmer these days absolutely needs to read and write. But a bronze age farmer? Not so much.

I personally think that one day it will be impossible to function in society without the ability to manipulate and query data. To use Jeff's example, as Mayor of a large city there must be mountains of data that might be relevant to every single decision that is made, and the mayors who can query and gain insight into that data quickly (without having to write a spec and get someone else to do it for them) will eventually be more successful, to the point where all mayors will be expected to be able to manipulate data.

There are a number of things that need to happen before that though:

Firstly, the tools need to get a LOT better. I put our current development tools somewhere a kin to Egyptian hieroglyphics. You can do basic things like record the births and deaths of kings, but could Shakespeare have written what he did in hieroglyphics? I'm guessing not. (Although I bet someone somewhere has done a translation!)

Next, this vision of the future is based on the amount of data in the world getting a LOT bigger. To the point where intuition or past experience simply can't compete with the people who are mining the data.

Politics will probably be one of the last professions to convert, the cultural change will take a while to work though. The first professions will be the kind of things where Excel is used heavily at the moment (you could argue that Excel is a programming environment). Professions like Marketing, Manufacturing (which will become predominately robot driven at some point) and Medicine (which will become heavily influenced by generic programming) will probably be early adopters. But one day, I think "programming" in the sense of giving a "machine" a series of logical instructions to follow, will be universal and it will simply be impossible to live a normal life without that skill.

EDIT: Fixed ironic typo ;)

I have a theory about bubbles: It's relatively easy to spot that you're in one, but it's very hard to pinpoint what sort of bubble it is. I'll explain...

During the lead up the 2008 debt crisis I saw a lot of people talking about how house prices had been going up year on year and questioning whether we were in a property bubble. There was a debate though. Demand for housing was strong (partly due to population growth) and it appeared that globalisation had freed up the capital so banks had plenty of money to lend to people to buy houses. So the rising price of property, it was argued, looked like the natural effect of market forces. In hindsight it's easy to see that there was an oversupply of wholesale debt, but at the time I think people assumed that it was just a more efficient distribution of global capital that was opening things up.

Similarly, in the lead up to 1999 people seemed to know that there was too much money floating around. Everyone seemed to be investing in stocks or property and swapping stories of how much they had made. Yes, tech stocks were rising quickly, but so was everything else, and there seemed to be no shortage of investors coming up with money to fund these tech companies so what's the problem.

I think what happens is that people always look at current circumstances in the context of recent bubbles and try to look for patterns. When they don't quite fit, they argue over whether this really is another instance of a previous bubble. My theory is that the overpriced asset is rarely the same thing as in the last 3 or 4 bubbles.

So, back to the present day. We have lots of high valuations for tech companies. But we also have lots of tech companies actually making a profit, plus operating costs are a lot lower. So we're not about to see a burst in the sense that the funding will dry up and companies will run out of runway. Instagram was like, 12 people(?), if they couldn't have got that $50mil investment I'm sure they could have found a way to keep operating for another year or two.

The challenge is to think about where the bubble might actually be. Could it be in advertising revenue? Are advertisers burning though reserves trying to get attention? Could it be in mineral resources? Are server or bandwidth costs artificially low? Could it be a bubble in intellectual property? Is the exact implementation of Facebook not actually as valuable as the companies valuation suggests. I have no idea, but I don't think this a complete re-run of 1999.

TLDR; I don't think this is "[Tech] Bubble 2.0", this is probably "[Something else] Bubble 1.0".

Just personal preference really, if your "isRelevant" filter was only ever going to be used by your "CountRelevantItems" method you might consider it more readable to have the logic of isRelevant right there in the CountRelevantItems method rather than elsewhere in the class.

I'm not sure there is one. They're useful in C# as a way of defining a function within another function (which sometimes makes for more readable code). For example:

    public int CountRelevantItems(IEnumerable<Thing> things)
    {
      // Non-trivial filter that you don't want in a where lamdba
      Func<Thing, bool> isRelevant = (t) =>
      {
        ...
      }
      
      return things.Where(t => isRelevant(t)).Count()
    }

But in Python you can already define functions inside of functions so you have a better solution:

    def CountRelevantThings(things):
      def isRelevant(thing):
        ...
      return len(filter(isRelevant, things))

I think that's possibly too simplistic. If the reason there's so much shit that needs doing is because the second guy didn't do it right first time then hiring the second guy might have been a mistake.

Maybe the first guy is just pissed that he has to put up with so much shit written by the second guy and is vocalising his desire to be free from it.

I prefer to think in terms of 'mean change turnaround' time; how quickly can your organisation or individual go from a problem or change request to functioning production system? If nothing ever breaks but new reports take 6 months, that's probably worse for business than the occasional outage and new reports taking 2 weeks.

Another way of looking at it would be that there's no way of proving that Earth was the starting point for all the life around us. So instead of looking for life that originated elsewhere we could find lots of "alien" life that looked remarkably similar to our own all over the galaxy.

If that life uses DNA, centuries from now the kind of genetic analysis we use to determine where one species branched of from another could be used to find where life on one planet came from.

Before the rail roads, people didn't do all that much travelling from A to B, unless they were very wealthy or had some external pressure like the need to find work or religious persecution, they mostly stayed at A. Which makes the invention of rail roads even more impressive in my eyes, it created demand rather than addressing it.

I disagree. She didn't have to have natural skin tones all over, and they didn't have to do a full frontal shot while she covered herself shyly, and the clothes they used didn't have to be so skimpy.

Creepy yes, I'm not sure about unrealistic, its a scifi story after all so I'm not sure what the benchmark for realism should be. But the whole sex bot thing and the nudity stuck me as unpleasant and unnecessary.

I think it's a question of balance, at the other end of the spectrum you can get growing startups where the initial "everyone's on a level and talks about everything" approach no longer works and has given way to a combination of multiple people competing over the same territory and the old guard finding ways to remind the noobs what the pecking order is. In that kind of situation, a little hierarchy and a few harmless ways of recognising past achievement/tenure would make for a much more harmonious work place.

Yes. We seem to be trying so hard these days to make our organisational structures appear flat and informal (e.g. open plan office spaces, casual dress policies, "360 degree reviews" and an aversion to using job titles) that I feel it's just pushed the competition for status and respect underground. The people who've been around for a long time simply find other ways to show newcomers that they're important or that they should be listened to. Things like camping out in one of the meeting rooms "because I have a lot of import calls to make". Or wearing something just slightly smarter than everyone else. Or a kind of faux humility like ScottGu's sub-title on his blog: http://weblogs.asp.net/scottgu/. Or coming to work in their 1998 "version 1.0 shipping date" T-shirt.

I've recently starting to think that people have a fundamental need for hierarchy and what you might call "respect structures" and that if you deny them the job titles and corner offices then they just find other ways of projecting their status. In a way, I'd prefer the overt methods - so long as there's a route to getting there myself someday.

In certain other countries the majority would find the "average American voters'" perspective on issues like economics or climate change quite strange. I can think of countries where there seems to be an almost universal acceptance of global warming but a poor acceptance of free market principles, for example.

My hypothesis is that in the long run education systems have a big impact on what average voters think; and that America's High School system isn't very strong on the scientific method. That's not to say it's all bad. I think it is very strong on creating a sense of national identity, something the UK system seems quite poor on.

    "Many forms of Government have been tried and will be
    tried in this world of sin and woe. No one pretends that 
    democracy is perfect or all-wise. Indeed, it has been
    said that democracy is the worst form of government 
    except all those other forms that have been tried from 
    time to time."

Churchill, 1947

http://hansard.millbanksystems.com/commons/1947/nov/11/parli...

I guess this sort of thing was bound to happen at some point. As the article says, all the major cloud platforms have had outages at some point.

As someone who is actively considering building products using this platform I'm keen to see how well they manage this issue. How do they communicate during the outage, how open about what went wrong are they afterwards, do they learn from it, and so on. I particularly like the efforts Amazon have gone to in the past to show they are learning from these issues (see http://aws.amazon.com/message/65648/) I'm hopeful that Microsoft will show the same level of openness.

The title made it seem like this was going to be more of an whistle-blowing blog post by an insider.

Nevertheless, this kind of rant is interesting in so far as it points to what I think a growing momentum behind the UNIX philosophy. What I mean by that is: small programs that do one thing well and interoperate with others.

I wonder whether the underlying cause of this shift in thinking is the levelling off of CPU cycle speeds. Time was when the performance of something like Photoshop or Office would just prompt you to buy a better computer. People would assume their computer had gotten old or out of date. Now that getting new hardware doesn't magically fix things people are asking why certain things take so long and comparing programs' performance.

This strikes me as no worse that the traditional approach of hiring recruiters and HR people to processes thousands of CVs and sit through hours of interviews.

The engineers have a route to distinguish themselves, they get well rewarded in the process, the big companies get the talent they want and society runs lots of experiments in the marketplace as a result.

Only only downside I can see is that a particular valuable service may get shut down, but at least the model has been proven and someone else in search of an idea could emulate it (think etherpad).

Exactly, the fundamentally hard bit of software development is the specification part.

As developers we translate between what the customer appears to want and what the compiler requires. If the customer had specified what they want with perfect clarity we would be able to automatically generate code from those requirements (or even compile them directly). But because they haven't been completely clear we have to use judgement and inference.

The fact that I will base my purchasing decisions on past experience of the brand and word of mouth about the product is not unusual. Many people will be like me, and this is one of the challenges that MS will have to work hard to overcome. Downvoting me for being that way is pretty immature; approach the world as it is, not how you want it to be.

I used a Windows Mobile for a couple of years (two different handsets). I found that while the main screens were reasonably well skinned and looked good in the store, when used for real that often fell away to reveal a clunky UI that was impossible to work with. I also found that I had to reboot every day because of lockups and crashes, plus the phone often dropped calls and was generally very slow.

I understand that this is a complete re-write of their phone OS, but I will still be waiting for a couple of years to see how others find it before I'm willing to trust that division of Microsoft again.

EDIT: I'll give you an example, our CTO recently got one of the new Windows Phones, I asked him what it was like and his response was: "Good... but I can't seem to get it connected to our Exchange server". If a rock star techie can't get his phone talking to Exchange what hope is there for the non-techies out there.

This is good news for Microsoft, and I really do hope they do well, but there's still a mountain to climb: tech reviewers are not the same as normal consumers and impressing them for the limited period of time it takes to write a review does not necessarily translate onto having a successful product. Here are a few challenges Microsoft still face:

1. Getting the mobile networks to take handsets running Windows Phone.

2. Training handset retailers on this new platform and promoting it enough to ensure that these handsets are actively sold to consumers.

3. Their advertising and PR must not be lame, like Microsoft advertising sometimes is.

4. The UX must be consistently slick enough over a long period of time that word of mouth among non-technical people really takes hold and the phone moves beyond the early adopter segment of the market. Phones must run for several days without a reboot. Calls must not be dropped because of crashes. UI Lockups or moments where the slick facade falls away to reveal clunky technical menus are a big no no. And the inevitable bugs must be fixed and rolled out in a timely and seamless mannor. Otherwise, when Bob from accounts notices you have a Windows Phone and asks "what's it like?" he'll sense the hesitation in your voice and choose the iPhone.

5. If things start to work out, the team will need top level executive cover to prevent in-fighting or politics from ripping the team appart or forcing an unwanted technical strategy tax (think "Windows Phone - Bing/Azure/Office Edition") from destroying the product.

6. Microsoft need to make money from these things, otherwise the project will be shelved and resourced diverted to MS Office 2020.