I've had the "pleasure" of using Google Docs when organisations I've worked with adopted it as their corporate standard.
Google Docs competes with Word in about the same way that a basic text editor with a distributed version control system does, except that Google Docs is slightly more interactive and much more likely to corrupt your document.
The red and green squiggly lines drove me to stop using the provided spelling and grammar checks in Word almost completely. It turns out that I misspell few enough words that a red-squiggly is far more likely to be Word either not knowing a word or assuming I'm in the US even though I'm using British English than it is to be a spelling error on my part. The green-squigglies have been a joke from the beginning and remain so to this day. I'll typically do a quick final check before I send a document to anyone else, which usually finds no genuine errors but does spot the occasional typo or doubled word from time to time, but that's an on-demand feature that doesn't require the use of anything squiggly.
That's an old argument in favour of typesetting systems like TeX and friends.
The thing is — and I write this as a guy who now produces serious documents directly in InDesign — it doesn't actually seem to disrupt much of anything in practice. InDesign uses a TeX-style H&J algorithm, so things do occasionally jump around slightly as you're editing. Since you're concentrating on typing, more often than not at the end of a paragraph if you're typing very much, and the layout effectively updates instantly on any modern computer, it's all quite stable by the time you pause to review what you just typed. Ditto for anchored frames that shift up and down a sidebar as you edit the main text, for example.
Of course, if you're working on a "chaotic" layout where typing a single extra character really can trigger a huge change like shifting a half-page float from the bottom of page 10 to the top of page 11, a perfectly sensible alternative is to have a dedicated story editor that is WYSIWYG up to a point but then display full page layout reflows in a separate window/panel/something.
The sad thing is that there are so many horrendous usability issues and functional limitations with both packages that these comparisons feel a bit like asking whether the guy hurtling to the ground at 98% of terminal velocity is in more trouble than the guy who reaches 98.5%, without anyone considering that giving either a parachute would have made a lot more difference.
I can't understand how market forces haven't yet driven anyone to create a general document editor that actually caters to what modern users want to do. Developing a program on this kind of scale is a serious project, and taking on Microsoft front and centre is probably a fool's game, but it's not as if you'd need a 500-strong team for years to do vastly better than any current word processor or mainstream DTP package, and Not Being Word isn't the commercial suicide note it used to be now that the dominant applications are the ability to (a) collaborate effectively within an organisation and (b) produce PDFs or other mostly "final" representations for external distribution.
Can't we Kickstarter enough funding for a small team of smart people with the right mix of skills to spend a year building something actually good?
Yes, everyone hates the ribbon interface! That's why Office 2010 has sold over 200 million copies. You'd think if it was so universally reviled and killed productivity (as slashdot claims with no proof), people would have stopped buying Office at 2007.
Your underlying point has merit, but your logic is also flawed.
For one thing, Slashdot, like the Office user base, is not a single person with a single mind. Different people have different preferences. In particular, most users of most software products are not so-called power users. The Ribbon interface works well for people who are not power users, and most such people do seem to prefer it once they get used to it, as Microsoft's usage data suggested they should.
However, that does not mean that the significant subset of Office users who really do intimately understand their way around a tried and tested combination of keyboard shortcuts, toolbar icons, menu commands, dialog tricks and so on will appreciate having the new UI and the underlying models forced on them as well. The Ribbon caters very much to cosmetic hacks and a quick-and-dirty approach. Don't bother defining styles, structuring your document systematically, or understanding how to present your data effectively! Just slap the format with the most "clever" borders on every table, format paint your headings so they're all the same colour that is a bit like your corporate standard, and use some random combination of bold, italic, faked small caps, underlining, colours, background colours and all-caps if you want to emphasize something. Oh, and if the spacing's not quite right, just hit enter a couple more times. Of course, MS Office has been going down this path for a long time and has never been shy about who it was aiming at, but the emphasis on the Ribbon pretty much seals its fate as any sort of productive tool for power users.
As for your 200 million sold copies statistic: the overwhelming majority of people who use MS Office do so because it came with their computer, it's their corporate standard at work, or it's the only thing they ever heard of so they pirated it. Microsoft sells about three individual licences a decade for Office applications and about a bazillion copies through mass licensing or preinstallation deals every year. The number of sales really doesn't tell us anything meaningful about what the people using Office actually think of the new ribbon.
Unfortunately, about 99.997% of documents written today are not academic papers or theses written to comply with the house style of a few hundred journals and a few hundred major institutions.
Even if they were, LaTeX's typesetting power now looks like the first car with an internal combustion engine: a revolutionary advance in technology at the time, that is now so antiquated and incompatible with modern standards that it has little value outside of its niche except as a historical curiosity.
Your argument about LaTeX controlling the logical design is well-taken, but unfortunately it never really did that, because in practice it conflated content and presentation to such an extent that you couldn't really separate them in anything beyond trivial cases.
The TeX family remains the preeminent tool for exactly one task today: typesetting maths. And that's only because no-one else has yet created another set of tools and fonts for doing so that doesn't suck.
I suspect there are quite a few of us out there who are sick of this, and I completely agree with pretty much everything you wrote.
I used to be a keen gamer. I have far more disposable income now, in my 30s, than I had say ten years ago when I would buy the big name titles, and I would happily spend some of it on good games today. However, I haven't bought a new AAA title in several years, for a few simple but almost universally applicable reasons:
1. If your game is so buggy that there's no guarantee that I will even be able to play it without frequent crashes, never mind actually complete it, I'm not going to start it. (That covers 100% of AAA titles I bought in the last year or two that I still bothered. There was nothing about my PCs that was in any way unusual or eccentric for a gaming machine of their respective generations, many of the bugs were widely reported, and quite a few were never fixed.)
2. If your game comes with software that acts like malware, even in the name of fighting piracy or getting rid of cheaters, I'm not going to install it on my PC any more than I would voluntarily install a virus or trojan horse. (And if you don't give me a cast-iron, no-nonsense statement of exactly what shady things you do, I'm going to assume your game includes malware these days. 100% of the AAA titles I didn't buy around the time I gave up would have failed on this criterion alone.)
3. If your game is incomplete, I'm not going to bother. A serious expansion pack at a fair price a few months after release is absolutely fine. That's a time-honoured way to extend the enjoyment of a good game and increase the returns for people who make good games, and I have no problem with any of that. Likewise, encouraging a modding community or giving away the odd extra freebie is all fine and good. These things all build on the original game and make it more worthwhile for everyone. But having different content available on day one depending on nothing but where you bought the game, or using paid-for DLC to fill in gaps in the main storyline or even adding DLC ads into the actual game, that's frankly just insulting.
4. Somewhat connected to point 1, I got fed up of the upgrade treadmill, having to retire an insanely powerful computer every year or two and replace it with the new shiny just to run a couple of new games. The time required to reinstall all the other stuff I use (particularly since I work from home and used the same PC for freelance work for a while) is prohibitive.
5. I like single-player games, which I can enjoy at my own pace and in my own time. Not everything has to be a multiplayer something-or-other in an endless open world.
And of course, the most important of all:
6. If your gameplay sucks, no amount of shiny and surround sound will save it. My favourite games from the moderately recent past are things like the Baldur's Gate series and Deus Ex. The kind of immersion you get in a deep, well-scripted storyline and somewhat open world mechanics are always going to beat any procedurally generated world with procedurally generated encounters and just a light dusting of actual human thought behind the story (I'm looking at you, Oblivion). My favourite FPS of all time, for both single player and multipler, is still Quake. My favourite RTS is still Total Annihilation, though Supreme Commander: Forged Alliance would have stolen the title if it hadn't been so damned buggy.
The sad thing is, it seems like there have been at least a few games in recent years that are in the spirit of the originals and obviously with better production values as well, but every time I see one all the DLC and malware and other such silliness just puts me off it. For this reason, I have also recently signed up for GOG. I have high hopes for the Baldur's Gate update, too, not just for being a BG remake, but because I'm hoping it will show that a team who actually seem to care more about making a good RPG th
The abuse wasn't lifting the fingerprints, it was accessing a private database containing sensitive information about members of the public for reasons other than those that were properly authorised.
Is it really an abuse when the only legal avenue you have to address such situations is completely uninterested in doing anything to help you?
Yes, I think it is. As my father taught me, two wrongs do not make a right.
What was going to come of abusing the ability to access the database in that way? What exactly did they propose to do if and when they identified the person responsible? They presumably can't take legitimate legal action, if their evidence was obtained illegally and becomes inadmissible as a result. <sarcasm> They might know where the guy lives, so I suppose they could just go round to his place, where knee-capping him will teach him the error of his ways. </sarcasm> Nothing good came of their action. All it showed was that someone in a position of public trust is not responsible enough to hold that position.
Of course, the fact that the police didn't do their job stinks. No-one is disputing that. But vigilantism is rarely a useful solution to anything, and when it is, we're usually talking more on the scale of toppling an entire government. If the victims here really wanted to improve the situation, couldn't they have collected the fingerprints (legally) and then gone to the media to show how easy it was and highlight the failure of the police to do their job properly? At least that way, even if their own case couldn't benefit, it would highlight the problem and incentivise those responsible to do better next time.
I was arguing that monitoring doesn't need to be "as you suggest", and that the biggest problem with monitoring "as you suggest" is the one-sidedness of the process and not the fact that the data existed in the first place. As I tried to demonstrate, the data itself could be just as valuable in this case when used by the defence in a proper trial as part of due process. That's a long way from supporting real-time access by outside parties and, for example, inviting fishing expeditions.
So you support the government monitoring you, because you are doing nothing wrong?
No, I don't. Reading my posting history, or even just glancing at my sig, should tell you how unlikely that is.
I'm simply pointing out that if there were objective data available, which presumably someone could present in their own defence at trial, then that would be a more robust way of protecting more innocent people than assuming that everyone can produce the kind of mathematical/scientific argument that this particular person did.
I don't know how you turned that into some sort of state surveillance argument, because I never mentioned any such thing. I will suggest to you, however, that a knee-jerk association of any kind of recording device with a surveillance state is probably not going to help when it comes to opposing actually over-reaching state surveillance.
While I agree with you about inappropriate policing priorities, I have to say I find your actions in abusing privileged access to a database to be more offensive than those of the person who broke into your car. Your friend at the DCJS who committed that "minor violation" should have been fired and prosecuted. The fact that you were right in the end does not justify the means, and we must never allow that kind of rationalisation to excuse abuse of public trust when officials have access to sensitive personal information.
At the risk of stating the obvious, if the defendant in this case had had a black box installed that could provide an accurate indication of his actions, he wouldn't have needed all that science (which of course many defendants who were similarly innocent could not have produced) to refute the officer's mistaken allegation.
Observations and facts are fine, it's one-sided observations and asymmetric access to analytic resources that tend to screw things up.
Many of the interesting things Facebook knows (or thinks it knows) about you don't come from what you told them yourself. They have such comprehensive surveillance of so much of society because other people are volunteering information that they can paint a pretty good picture of you just from a couple of basic facts you might give them thinking they don't mean much and the vast networking effects that they can data mine.
For example, suppose you only have a Facebook account with a couple of basic personal details, and don't post updates, photos, etc. yourself. However, assume you are Facebook friends with your real life friends so you can follow their news, get invited to their parties, etc. One day, Bill sends you and five other friends an invitation to a social event at a certain place and time. You don't respond on Facebook, but the others all do, and six of you (one couldn't make it) go enjoy a drink after work.
Three weeks later, suppose someone completely unrelated to you or your friends posts a photo they took in the bar that night, which has the six of you in the background playing pool. Facebook can potentially guess the approximate time and place of the photo from metadata (or just because the other person told them) and match it to your friend's invitation. Facebook might also have photos of your five companions, and you are the sixth guy around the table. It's now a pretty good bet that Facebook has a photo of you that they can or will credibly be able to identify, even though you did nothing but receive an invitation from a friend. None of your own friends, who maybe know that you don't like putting too much info on Facebook, even had to provide the photo or tag you in it. Obviously Facebook can also guess that you did in fact attend the event you were invited to, even though you didn't tell them that.
This isn't meant to describe a specific practice that is necessarily happening today, and it's just the first example that came to mind. However, all the necessary technologies are at least working well in R&D labs, and obviously Facebook routinely data mine all kinds of network relationships, so the sort of situation I described isn't exactly a huge stretch. Also, notice how cagey Facebook have been about disclosing exactly what personal data they hold about people, even when required to do so under European law: they seem to be relying on legal weasel words to try to exempt part of the data they have on people from disclosure, on the basis that it would give away their trade secrets or some similar argument. That's a pretty strong hint that they are doing some things with the data that they aren't publicly disclosing and that don't involve just one individual's freely given information.
His point is that society is collectively tweaking the norm of what is acceptable, and the police and politicians are exploiting this.
Is society really "tweaking the norms" all that much? It seems quite likely that the kind of person who posts a lot of detailed updates on Facebook or Twitter doesn't value privacy as highly. It also seems quite likely that such people will be seen/read more often on-line than those more private individuals with dissenting views. Assuming that reduced privacy is the new social norm because the balance of on-line commentary says so is a classic case of confirmation bias.
Privacy is a particularly dangerous area to make such assumptions anyway, partly because of the inherent Pandora's box effect, and partly because so many people don't actually understand how much of their privacy is being surrendered when they choose to use certain services. There have been plenty of cases where loads of people used a system, yet when presented with the facts about the privacy implications, their views then became quite hostile toward that system.
Personally, I don't have a Facebook account, post my life story on Twitter, or discuss my private medical conditions on a crowded train.
When I worked for someone else, I accepted that the company could in theory monitor communications I sent from company systems. However, (a) they were company systems paid for by the company and provided for work, (b) I was clearly told that this was a possibility, and (c) the major reason for them spending the money on the people and equipment who might perform that monitoring was compliance with legal obligations in various countries. Any employer is likely to be in a catch-22 situation with modern laws in most western countries on this one, even if they have nothing but respect for their employees' privacy.
In short, I do not voluntarily give up my privacy in the kinds of ways that this lawyer describes, and when it comes to another party invading my privacy, I don't consider the willingness of other people to give up their own privacy to be any sort of justification. It is more than a little ironic that in a discussion about privacy, of all things, someone should be making an argument that fundamentally assumes everyone thinks and acts the same way.
Well, it's clear that you don't know much about either basic economics or the reality of running a business that sells electronic goods. There is probably little point trying to debate the points with you in that case, but I'll repond to a couple of them all the same. Go start a couple of businesses that actually do this, see what really costs what and how you can make the business viable, and then maybe we can talk more seriously.
I will say that on "fleecing the rich", rather like progressive taxation, the reason you charge more to those who can afford it for high-value works that are expensive to produce is that if you charged the same amount to everyone, most people couldn't or wouldn't pay and your product might literally not be financially viable. The marginal costs of distribution are low for digital media, but you still have to make back the money that paid to create the work the first place, and make a reasonable profit, and you need to know ahead of time that there is a likelihood of making enough profit on some works to offset the sunk costs of the others if they prove not to be popular or you're not going to invest in creating anything commercially at all.
There is nothing "rip-off" about this. It is simple economics. For one thing, it is unlikely that anyone is forced to buy any given electronic work, so competitive pressure will tend to avoid prices becoming unreasonable to the consumer. For another, you may have become conditioned to expect that digital media will be either low-cost if bought directly or free if you rip them off, but that may or may not have anything to do with the actual costs of producing and supplying those works, which might not be relevant to you but obviously are to a business. Most of us aren't making this work out of charity, so if you want to dramatically reduce the benefits of copyright as an incentive, suck it up and understand that we will need to maximise profits over the short term for which copyright does apply or we're not going to play your little game at all. That's why copyright exists as an incentive in the first place, and it's why it has to remain a clear enough benefit to those who create and distribute works.
And as a copyright holder, you should have to make your media available to anyone on equal terms and in (applicable to the type of media) standard forms at a price that can stay the same (adjusted for inflation) or go down, but not go up.
Sure, just as soon as you arrange consistent taxation rules across the entire world so it costs me the same to sell each copy wherever my customer is. Oh, and you have to enforce copyright reliably so that I can pitch my work at a price that maximizes my profits within your permitted five-year period based on wealthy markets who can afford to pay, without everyone else just ripping me off. Also, every student in the world who might have enjoyed the work at a discount because I was a student once myself now hates you. BTW, you didn't want to ever buy anything interesting at a somewhat reduced price during an economic downturn, did you? Don't forget you also need to provide a guaranteed, cost-free, universal sales channel throughout the world so that I can sell my work to anyone on the planet from day one and keep it available for the full five years to avoid inadvertently giving up my copyright.
There are far too many works out there which will be completely forgotten by the time the copyright on them expires, there may not even be any readable copies left by that time resulting in the loss of that work.
That is debatable on all counts.
Also, there are far too many people who have developed or even grown up with a distorted sense of entitlement, in whose world good artistic works magically grow on trees and it isn't necessary to actually pay all the people who worked hard to create them. Such people tend to have a rather limited and distorted understanding of economics, and consequently they propose naive, simplistic ideas without really thinking through the likely consequences. Your version isn't as bad as the "just abolish copyright" crowd, but I still think it's broken in far too many ways to be credible.
Copyright is not only about revenue and money people, it is also about control of the actual use of your work, you do not want your music to appear in some gay porn video
Actually, no it isn't. In fact, in many jurisdictions, there are separate laws to cover various moral issues such as recognising the original artist and not using the work in some inappropriate way that reflects badly on the artist as a result. That allows the freedom to determine copyright rules purely on economic merit, though of course whether the relevant legislature are willing and/or able to do so is a different question.
It's actually more complicated even than that, though, because there may be different VAT rates for different kinds of product or service, some of which may be 0%, which in turn isn't necessarily the same thing as being exempt. So then we have the European Parliament resolution from the end of last year on lowering the tax rate for "digital cultural goods" like e-books, because these are currently considered services and taxed accordingly even though paper books attract a reduced or zero rate in many EU countries.
Also, the rules are expected to change in the fairly near future so that sales of electronic goods within the EU would attract VAT or equivalent sales tax at the purchaser's home rate, which will defeat the Luxembourg trick.
Basically, it's all a horrendous mess and the goalposts are constantly shifting.:-(
FYI, the rules on sales tax (VAT etc.) in Europe are quite a touchy subject, in particular whether distance selling customers pay their local rate or the rate where the vendor is based. Right now, it's possible for sales to be charged at the vendor's local rate even if the purchaser is in another country with a higher rate. This is likely to change within the next couple of years, though, to avoid the kind of tax dodge you described.
As someone setting up a business to sell on-line, I'm in two minds about this. On the one hand, I'm in the UK, where VAT is relatively high compared to much of Europe. The price my customers pay would therefore drop in some cases, making my product more attractive.
On the other hand, for a small business, dealing with taxes is already a significant overhead in both time and, probably fees paid to an accountant. If everyone selling on-line has to update their web stores (a) to handle different sales tax rates based on where the buyer is, instead of the simple yes/no that typically applies today, and (b) to track the current sales tax rates from day to day in every country affected, then this is a major overhaul that's going to hurt a lot of people using fairly simple and often bought-in e-commerce platforms.
You're probably better off reading some of the published work on the subject than relying on any small number of anecdotes I can remember off the top of my head right now. A lot of my background comes from an extended training/process improvement exercise I took part in, but unfortunately as I'm no longer in the same role I've now handed on or filed away most of the detailed reference material from that period.
If you have a copy of Code Complete, that's an obvious place you could start, because there's definitely a section in there that cites some surveys, which in turn cite plenty of real world case studies with rather consistent results. If memory serves, McConnell also gives separate statistics on the effectiveness of code reviews and design reviews, and compares them with other techniques such as unit testing, beta testing, and so on.
If you want something more detailed, there are also books dedicated entirely to software testing or even to peer review specifically. These typically cite plenty more studies with hard data to back up their case. Jason Cohen and a few others published a collection of essays about peer review in 2006, and probably have more material since then given that A Smart Bear makes software to aid in performing reviews. Karl Wiegers wrote a book specifically on reviews somewhere around 2002. Ed Kit's company SDT gives dedicated technical review training and cited plenty of sources when we worked with them, so anything on the subject that they are publishing these days is probably relevant.
An obvious example is doing technical reviews throughout the development process. That includes code reviews, but also earlier stages like checking that requirements are understood up-front and checking that a proposed design strategy is reasonable.
A good peer review process identifies potential bugs earlier, when they are easier and cheaper to prevent. Based on empirical data from real world studies, we know a systematic review proces can cut the number of bugs that escape into production by as much as an order of magnitude. Typically, it also saves a substantial amount of time and money, because correcting bugs in production is orders of magnitude more expensive than spotting an ambiguous requirement or design flaw in the early stages or at least catching a bug before it gets in front of customers.
However, when you start mentioning code reviews, a lot of developers who've been around the block a few times immediately envisage a heavyweight, Fagan-style review process where life becomes dominated by long, tiresome meetings. These developers may become hostile as soon as the words "code review" are even mentioned, without even knowing about modern processes and tools.
More recently, some developers favour Agile processes with very rapid release cycles, sometimes pushing code into production several times a day, and perhaps TDD. These process elements naturally conflict with a peer review process of the kind I described, and so the best you are likely to see is pair programming or a token second-pair-of-eyes glance over code before it's merged into production. While better than no review at all, such processes are nowhere near as effective as a structured peer review.
Of course, I'm only concentrating on one possible process improvement in this post. There are plenty more ideas that could help many projects but aren't nearly as widely used as they could be, ranging from relatively simple things like using automated test suites to much heavier things like formal methods.
Moreover I'm sticking to processes here because that's what you asked about, but of course there are also many different programming languages and related tools that inherently close off entire attack vectors left wide open by a lot of software today, hence my criticism of using C or C++ for most security-sensitive work.
In the Internet age, where networking code and communication clients might be used by millions of people, and where a single exploit might therefore lead to downtime, data leakage or becoming part of a botnet for millions of people, is it so unreasonable to expect that software like web browsers and e-mail clients should be written by something a little more advanced than glorified trial-and-error?
As far as look and feel goes, it's a fallacy that all you have to do to make a GUI feel native is change the chrome. You can't turn a Windows native application into an OS X native application just by altering how you draw a button and a checkbox. No GUI toolkit that is based on the assumption that you can will ever be any good for making professional level user interfaces on any platform (or at least, on any more than one platform that it secretly favours).
There are two ways of constructing a software design.
One way is to make it so simple that there are obviously no deficiencies, and the other way is to make it so complicated that there are no obvious deficiencies.
Slashdot Mirror
I've had the "pleasure" of using Google Docs when organisations I've worked with adopted it as their corporate standard.
Google Docs competes with Word in about the same way that a basic text editor with a distributed version control system does, except that Google Docs is slightly more interactive and much more likely to corrupt your document.
The red and green squiggly lines drove me to stop using the provided spelling and grammar checks in Word almost completely. It turns out that I misspell few enough words that a red-squiggly is far more likely to be Word either not knowing a word or assuming I'm in the US even though I'm using British English than it is to be a spelling error on my part. The green-squigglies have been a joke from the beginning and remain so to this day. I'll typically do a quick final check before I send a document to anyone else, which usually finds no genuine errors but does spot the occasional typo or doubled word from time to time, but that's an on-demand feature that doesn't require the use of anything squiggly.
That's an old argument in favour of typesetting systems like TeX and friends.
The thing is — and I write this as a guy who now produces serious documents directly in InDesign — it doesn't actually seem to disrupt much of anything in practice. InDesign uses a TeX-style H&J algorithm, so things do occasionally jump around slightly as you're editing. Since you're concentrating on typing, more often than not at the end of a paragraph if you're typing very much, and the layout effectively updates instantly on any modern computer, it's all quite stable by the time you pause to review what you just typed. Ditto for anchored frames that shift up and down a sidebar as you edit the main text, for example.
Of course, if you're working on a "chaotic" layout where typing a single extra character really can trigger a huge change like shifting a half-page float from the bottom of page 10 to the top of page 11, a perfectly sensible alternative is to have a dedicated story editor that is WYSIWYG up to a point but then display full page layout reflows in a separate window/panel/something.
The sad thing is that there are so many horrendous usability issues and functional limitations with both packages that these comparisons feel a bit like asking whether the guy hurtling to the ground at 98% of terminal velocity is in more trouble than the guy who reaches 98.5%, without anyone considering that giving either a parachute would have made a lot more difference.
I can't understand how market forces haven't yet driven anyone to create a general document editor that actually caters to what modern users want to do. Developing a program on this kind of scale is a serious project, and taking on Microsoft front and centre is probably a fool's game, but it's not as if you'd need a 500-strong team for years to do vastly better than any current word processor or mainstream DTP package, and Not Being Word isn't the commercial suicide note it used to be now that the dominant applications are the ability to (a) collaborate effectively within an organisation and (b) produce PDFs or other mostly "final" representations for external distribution.
Can't we Kickstarter enough funding for a small team of smart people with the right mix of skills to spend a year building something actually good?
Yes, everyone hates the ribbon interface! That's why Office 2010 has sold over 200 million copies. You'd think if it was so universally reviled and killed productivity (as slashdot claims with no proof), people would have stopped buying Office at 2007.
Your underlying point has merit, but your logic is also flawed.
For one thing, Slashdot, like the Office user base, is not a single person with a single mind. Different people have different preferences. In particular, most users of most software products are not so-called power users. The Ribbon interface works well for people who are not power users, and most such people do seem to prefer it once they get used to it, as Microsoft's usage data suggested they should.
However, that does not mean that the significant subset of Office users who really do intimately understand their way around a tried and tested combination of keyboard shortcuts, toolbar icons, menu commands, dialog tricks and so on will appreciate having the new UI and the underlying models forced on them as well. The Ribbon caters very much to cosmetic hacks and a quick-and-dirty approach. Don't bother defining styles, structuring your document systematically, or understanding how to present your data effectively! Just slap the format with the most "clever" borders on every table, format paint your headings so they're all the same colour that is a bit like your corporate standard, and use some random combination of bold, italic, faked small caps, underlining, colours, background colours and all-caps if you want to emphasize something. Oh, and if the spacing's not quite right, just hit enter a couple more times. Of course, MS Office has been going down this path for a long time and has never been shy about who it was aiming at, but the emphasis on the Ribbon pretty much seals its fate as any sort of productive tool for power users.
As for your 200 million sold copies statistic: the overwhelming majority of people who use MS Office do so because it came with their computer, it's their corporate standard at work, or it's the only thing they ever heard of so they pirated it. Microsoft sells about three individual licences a decade for Office applications and about a bazillion copies through mass licensing or preinstallation deals every year. The number of sales really doesn't tell us anything meaningful about what the people using Office actually think of the new ribbon.
Unfortunately, about 99.997% of documents written today are not academic papers or theses written to comply with the house style of a few hundred journals and a few hundred major institutions.
Even if they were, LaTeX's typesetting power now looks like the first car with an internal combustion engine: a revolutionary advance in technology at the time, that is now so antiquated and incompatible with modern standards that it has little value outside of its niche except as a historical curiosity.
Your argument about LaTeX controlling the logical design is well-taken, but unfortunately it never really did that, because in practice it conflated content and presentation to such an extent that you couldn't really separate them in anything beyond trivial cases.
The TeX family remains the preeminent tool for exactly one task today: typesetting maths. And that's only because no-one else has yet created another set of tools and fonts for doing so that doesn't suck.
I suspect there are quite a few of us out there who are sick of this, and I completely agree with pretty much everything you wrote.
I used to be a keen gamer. I have far more disposable income now, in my 30s, than I had say ten years ago when I would buy the big name titles, and I would happily spend some of it on good games today. However, I haven't bought a new AAA title in several years, for a few simple but almost universally applicable reasons:
1. If your game is so buggy that there's no guarantee that I will even be able to play it without frequent crashes, never mind actually complete it, I'm not going to start it. (That covers 100% of AAA titles I bought in the last year or two that I still bothered. There was nothing about my PCs that was in any way unusual or eccentric for a gaming machine of their respective generations, many of the bugs were widely reported, and quite a few were never fixed.)
2. If your game comes with software that acts like malware, even in the name of fighting piracy or getting rid of cheaters, I'm not going to install it on my PC any more than I would voluntarily install a virus or trojan horse. (And if you don't give me a cast-iron, no-nonsense statement of exactly what shady things you do, I'm going to assume your game includes malware these days. 100% of the AAA titles I didn't buy around the time I gave up would have failed on this criterion alone.)
3. If your game is incomplete, I'm not going to bother. A serious expansion pack at a fair price a few months after release is absolutely fine. That's a time-honoured way to extend the enjoyment of a good game and increase the returns for people who make good games, and I have no problem with any of that. Likewise, encouraging a modding community or giving away the odd extra freebie is all fine and good. These things all build on the original game and make it more worthwhile for everyone. But having different content available on day one depending on nothing but where you bought the game, or using paid-for DLC to fill in gaps in the main storyline or even adding DLC ads into the actual game, that's frankly just insulting.
4. Somewhat connected to point 1, I got fed up of the upgrade treadmill, having to retire an insanely powerful computer every year or two and replace it with the new shiny just to run a couple of new games. The time required to reinstall all the other stuff I use (particularly since I work from home and used the same PC for freelance work for a while) is prohibitive.
5. I like single-player games, which I can enjoy at my own pace and in my own time. Not everything has to be a multiplayer something-or-other in an endless open world.
And of course, the most important of all:
6. If your gameplay sucks, no amount of shiny and surround sound will save it. My favourite games from the moderately recent past are things like the Baldur's Gate series and Deus Ex. The kind of immersion you get in a deep, well-scripted storyline and somewhat open world mechanics are always going to beat any procedurally generated world with procedurally generated encounters and just a light dusting of actual human thought behind the story (I'm looking at you, Oblivion). My favourite FPS of all time, for both single player and multipler, is still Quake. My favourite RTS is still Total Annihilation, though Supreme Commander: Forged Alliance would have stolen the title if it hadn't been so damned buggy.
The sad thing is, it seems like there have been at least a few games in recent years that are in the spirit of the originals and obviously with better production values as well, but every time I see one all the DLC and malware and other such silliness just puts me off it. For this reason, I have also recently signed up for GOG. I have high hopes for the Baldur's Gate update, too, not just for being a BG remake, but because I'm hoping it will show that a team who actually seem to care more about making a good RPG th
The abuse wasn't lifting the fingerprints, it was accessing a private database containing sensitive information about members of the public for reasons other than those that were properly authorised.
Is it really an abuse when the only legal avenue you have to address such situations is completely uninterested in doing anything to help you?
Yes, I think it is. As my father taught me, two wrongs do not make a right.
What was going to come of abusing the ability to access the database in that way? What exactly did they propose to do if and when they identified the person responsible? They presumably can't take legitimate legal action, if their evidence was obtained illegally and becomes inadmissible as a result. <sarcasm> They might know where the guy lives, so I suppose they could just go round to his place, where knee-capping him will teach him the error of his ways. </sarcasm> Nothing good came of their action. All it showed was that someone in a position of public trust is not responsible enough to hold that position.
Of course, the fact that the police didn't do their job stinks. No-one is disputing that. But vigilantism is rarely a useful solution to anything, and when it is, we're usually talking more on the scale of toppling an entire government. If the victims here really wanted to improve the situation, couldn't they have collected the fingerprints (legally) and then gone to the media to show how easy it was and highlight the failure of the police to do their job properly? At least that way, even if their own case couldn't benefit, it would highlight the problem and incentivise those responsible to do better next time.
I was arguing that monitoring doesn't need to be "as you suggest", and that the biggest problem with monitoring "as you suggest" is the one-sidedness of the process and not the fact that the data existed in the first place. As I tried to demonstrate, the data itself could be just as valuable in this case when used by the defence in a proper trial as part of due process. That's a long way from supporting real-time access by outside parties and, for example, inviting fishing expeditions.
So you support the government monitoring you, because you are doing nothing wrong?
No, I don't. Reading my posting history, or even just glancing at my sig, should tell you how unlikely that is.
I'm simply pointing out that if there were objective data available, which presumably someone could present in their own defence at trial, then that would be a more robust way of protecting more innocent people than assuming that everyone can produce the kind of mathematical/scientific argument that this particular person did.
I don't know how you turned that into some sort of state surveillance argument, because I never mentioned any such thing. I will suggest to you, however, that a knee-jerk association of any kind of recording device with a surveillance state is probably not going to help when it comes to opposing actually over-reaching state surveillance.
While I agree with you about inappropriate policing priorities, I have to say I find your actions in abusing privileged access to a database to be more offensive than those of the person who broke into your car. Your friend at the DCJS who committed that "minor violation" should have been fired and prosecuted. The fact that you were right in the end does not justify the means, and we must never allow that kind of rationalisation to excuse abuse of public trust when officials have access to sensitive personal information.
At the risk of stating the obvious, if the defendant in this case had had a black box installed that could provide an accurate indication of his actions, he wouldn't have needed all that science (which of course many defendants who were similarly innocent could not have produced) to refute the officer's mistaken allegation.
Observations and facts are fine, it's one-sided observations and asymmetric access to analytic resources that tend to screw things up.
Many of the interesting things Facebook knows (or thinks it knows) about you don't come from what you told them yourself. They have such comprehensive surveillance of so much of society because other people are volunteering information that they can paint a pretty good picture of you just from a couple of basic facts you might give them thinking they don't mean much and the vast networking effects that they can data mine.
For example, suppose you only have a Facebook account with a couple of basic personal details, and don't post updates, photos, etc. yourself. However, assume you are Facebook friends with your real life friends so you can follow their news, get invited to their parties, etc. One day, Bill sends you and five other friends an invitation to a social event at a certain place and time. You don't respond on Facebook, but the others all do, and six of you (one couldn't make it) go enjoy a drink after work.
Three weeks later, suppose someone completely unrelated to you or your friends posts a photo they took in the bar that night, which has the six of you in the background playing pool. Facebook can potentially guess the approximate time and place of the photo from metadata (or just because the other person told them) and match it to your friend's invitation. Facebook might also have photos of your five companions, and you are the sixth guy around the table. It's now a pretty good bet that Facebook has a photo of you that they can or will credibly be able to identify, even though you did nothing but receive an invitation from a friend. None of your own friends, who maybe know that you don't like putting too much info on Facebook, even had to provide the photo or tag you in it. Obviously Facebook can also guess that you did in fact attend the event you were invited to, even though you didn't tell them that.
This isn't meant to describe a specific practice that is necessarily happening today, and it's just the first example that came to mind. However, all the necessary technologies are at least working well in R&D labs, and obviously Facebook routinely data mine all kinds of network relationships, so the sort of situation I described isn't exactly a huge stretch. Also, notice how cagey Facebook have been about disclosing exactly what personal data they hold about people, even when required to do so under European law: they seem to be relying on legal weasel words to try to exempt part of the data they have on people from disclosure, on the basis that it would give away their trade secrets or some similar argument. That's a pretty strong hint that they are doing some things with the data that they aren't publicly disclosing and that don't involve just one individual's freely given information.
His point is that society is collectively tweaking the norm of what is acceptable, and the police and politicians are exploiting this.
Is society really "tweaking the norms" all that much? It seems quite likely that the kind of person who posts a lot of detailed updates on Facebook or Twitter doesn't value privacy as highly. It also seems quite likely that such people will be seen/read more often on-line than those more private individuals with dissenting views. Assuming that reduced privacy is the new social norm because the balance of on-line commentary says so is a classic case of confirmation bias.
Privacy is a particularly dangerous area to make such assumptions anyway, partly because of the inherent Pandora's box effect, and partly because so many people don't actually understand how much of their privacy is being surrendered when they choose to use certain services. There have been plenty of cases where loads of people used a system, yet when presented with the facts about the privacy implications, their views then became quite hostile toward that system.
That is my problem with the argument as well.
Personally, I don't have a Facebook account, post my life story on Twitter, or discuss my private medical conditions on a crowded train.
When I worked for someone else, I accepted that the company could in theory monitor communications I sent from company systems. However, (a) they were company systems paid for by the company and provided for work, (b) I was clearly told that this was a possibility, and (c) the major reason for them spending the money on the people and equipment who might perform that monitoring was compliance with legal obligations in various countries. Any employer is likely to be in a catch-22 situation with modern laws in most western countries on this one, even if they have nothing but respect for their employees' privacy.
In short, I do not voluntarily give up my privacy in the kinds of ways that this lawyer describes, and when it comes to another party invading my privacy, I don't consider the willingness of other people to give up their own privacy to be any sort of justification. It is more than a little ironic that in a discussion about privacy, of all things, someone should be making an argument that fundamentally assumes everyone thinks and acts the same way.
Well, it's clear that you don't know much about either basic economics or the reality of running a business that sells electronic goods. There is probably little point trying to debate the points with you in that case, but I'll repond to a couple of them all the same. Go start a couple of businesses that actually do this, see what really costs what and how you can make the business viable, and then maybe we can talk more seriously.
I will say that on "fleecing the rich", rather like progressive taxation, the reason you charge more to those who can afford it for high-value works that are expensive to produce is that if you charged the same amount to everyone, most people couldn't or wouldn't pay and your product might literally not be financially viable. The marginal costs of distribution are low for digital media, but you still have to make back the money that paid to create the work the first place, and make a reasonable profit, and you need to know ahead of time that there is a likelihood of making enough profit on some works to offset the sunk costs of the others if they prove not to be popular or you're not going to invest in creating anything commercially at all.
There is nothing "rip-off" about this. It is simple economics. For one thing, it is unlikely that anyone is forced to buy any given electronic work, so competitive pressure will tend to avoid prices becoming unreasonable to the consumer. For another, you may have become conditioned to expect that digital media will be either low-cost if bought directly or free if you rip them off, but that may or may not have anything to do with the actual costs of producing and supplying those works, which might not be relevant to you but obviously are to a business. Most of us aren't making this work out of charity, so if you want to dramatically reduce the benefits of copyright as an incentive, suck it up and understand that we will need to maximise profits over the short term for which copyright does apply or we're not going to play your little game at all. That's why copyright exists as an incentive in the first place, and it's why it has to remain a clear enough benefit to those who create and distribute works.
And as a copyright holder, you should have to make your media available to anyone on equal terms and in (applicable to the type of media) standard forms at a price that can stay the same (adjusted for inflation) or go down, but not go up.
Sure, just as soon as you arrange consistent taxation rules across the entire world so it costs me the same to sell each copy wherever my customer is. Oh, and you have to enforce copyright reliably so that I can pitch my work at a price that maximizes my profits within your permitted five-year period based on wealthy markets who can afford to pay, without everyone else just ripping me off. Also, every student in the world who might have enjoyed the work at a discount because I was a student once myself now hates you. BTW, you didn't want to ever buy anything interesting at a somewhat reduced price during an economic downturn, did you? Don't forget you also need to provide a guaranteed, cost-free, universal sales channel throughout the world so that I can sell my work to anyone on the planet from day one and keep it available for the full five years to avoid inadvertently giving up my copyright.
There are far too many works out there which will be completely forgotten by the time the copyright on them expires, there may not even be any readable copies left by that time resulting in the loss of that work.
That is debatable on all counts.
Also, there are far too many people who have developed or even grown up with a distorted sense of entitlement, in whose world good artistic works magically grow on trees and it isn't necessary to actually pay all the people who worked hard to create them. Such people tend to have a rather limited and distorted understanding of economics, and consequently they propose naive, simplistic ideas without really thinking through the likely consequences. Your version isn't as bad as the "just abolish copyright" crowd, but I still think it's broken in far too many ways to be credible.
Copyright is not only about revenue and money people, it is also about control of the actual use of your work, you do not want your music to appear in some gay porn video
Actually, no it isn't. In fact, in many jurisdictions, there are separate laws to cover various moral issues such as recognising the original artist and not using the work in some inappropriate way that reflects badly on the artist as a result. That allows the freedom to determine copyright rules purely on economic merit, though of course whether the relevant legislature are willing and/or able to do so is a different question.
It's actually more complicated even than that, though, because there may be different VAT rates for different kinds of product or service, some of which may be 0%, which in turn isn't necessarily the same thing as being exempt. So then we have the European Parliament resolution from the end of last year on lowering the tax rate for "digital cultural goods" like e-books, because these are currently considered services and taxed accordingly even though paper books attract a reduced or zero rate in many EU countries.
Also, the rules are expected to change in the fairly near future so that sales of electronic goods within the EU would attract VAT or equivalent sales tax at the purchaser's home rate, which will defeat the Luxembourg trick.
Basically, it's all a horrendous mess and the goalposts are constantly shifting. :-(
FYI, the rules on sales tax (VAT etc.) in Europe are quite a touchy subject, in particular whether distance selling customers pay their local rate or the rate where the vendor is based. Right now, it's possible for sales to be charged at the vendor's local rate even if the purchaser is in another country with a higher rate. This is likely to change within the next couple of years, though, to avoid the kind of tax dodge you described.
As someone setting up a business to sell on-line, I'm in two minds about this. On the one hand, I'm in the UK, where VAT is relatively high compared to much of Europe. The price my customers pay would therefore drop in some cases, making my product more attractive.
On the other hand, for a small business, dealing with taxes is already a significant overhead in both time and, probably fees paid to an accountant. If everyone selling on-line has to update their web stores (a) to handle different sales tax rates based on where the buyer is, instead of the simple yes/no that typically applies today, and (b) to track the current sales tax rates from day to day in every country affected, then this is a major overhaul that's going to hurt a lot of people using fairly simple and often bought-in e-commerce platforms.
You're probably better off reading some of the published work on the subject than relying on any small number of anecdotes I can remember off the top of my head right now. A lot of my background comes from an extended training/process improvement exercise I took part in, but unfortunately as I'm no longer in the same role I've now handed on or filed away most of the detailed reference material from that period.
If you have a copy of Code Complete, that's an obvious place you could start, because there's definitely a section in there that cites some surveys, which in turn cite plenty of real world case studies with rather consistent results. If memory serves, McConnell also gives separate statistics on the effectiveness of code reviews and design reviews, and compares them with other techniques such as unit testing, beta testing, and so on.
If you want something more detailed, there are also books dedicated entirely to software testing or even to peer review specifically. These typically cite plenty more studies with hard data to back up their case. Jason Cohen and a few others published a collection of essays about peer review in 2006, and probably have more material since then given that A Smart Bear makes software to aid in performing reviews. Karl Wiegers wrote a book specifically on reviews somewhere around 2002. Ed Kit's company SDT gives dedicated technical review training and cited plenty of sources when we worked with them, so anything on the subject that they are publishing these days is probably relevant.
An obvious example is doing technical reviews throughout the development process. That includes code reviews, but also earlier stages like checking that requirements are understood up-front and checking that a proposed design strategy is reasonable.
A good peer review process identifies potential bugs earlier, when they are easier and cheaper to prevent. Based on empirical data from real world studies, we know a systematic review proces can cut the number of bugs that escape into production by as much as an order of magnitude. Typically, it also saves a substantial amount of time and money, because correcting bugs in production is orders of magnitude more expensive than spotting an ambiguous requirement or design flaw in the early stages or at least catching a bug before it gets in front of customers.
However, when you start mentioning code reviews, a lot of developers who've been around the block a few times immediately envisage a heavyweight, Fagan-style review process where life becomes dominated by long, tiresome meetings. These developers may become hostile as soon as the words "code review" are even mentioned, without even knowing about modern processes and tools.
More recently, some developers favour Agile processes with very rapid release cycles, sometimes pushing code into production several times a day, and perhaps TDD. These process elements naturally conflict with a peer review process of the kind I described, and so the best you are likely to see is pair programming or a token second-pair-of-eyes glance over code before it's merged into production. While better than no review at all, such processes are nowhere near as effective as a structured peer review.
Of course, I'm only concentrating on one possible process improvement in this post. There are plenty more ideas that could help many projects but aren't nearly as widely used as they could be, ranging from relatively simple things like using automated test suites to much heavier things like formal methods.
Moreover I'm sticking to processes here because that's what you asked about, but of course there are also many different programming languages and related tools that inherently close off entire attack vectors left wide open by a lot of software today, hence my criticism of using C or C++ for most security-sensitive work.
In the Internet age, where networking code and communication clients might be used by millions of people, and where a single exploit might therefore lead to downtime, data leakage or becoming part of a botnet for millions of people, is it so unreasonable to expect that software like web browsers and e-mail clients should be written by something a little more advanced than glorified trial-and-error?
As far as look and feel goes, it's a fallacy that all you have to do to make a GUI feel native is change the chrome. You can't turn a Windows native application into an OS X native application just by altering how you draw a button and a checkbox. No GUI toolkit that is based on the assumption that you can will ever be any good for making professional level user interfaces on any platform (or at least, on any more than one platform that it secretly favours).
There are two ways of constructing a software design.
One way is to make it so simple that there are obviously no deficiencies, and the other way is to make it so complicated that there are no obvious deficiencies.
The first method is far more difficult.
C. A. R. Hoare, 1980 Turing Award lecture