Microsoft is holding back the secure web
blog.jgc.org
blog.jgc.org
And I strongly disagree with the title that MS is the one holding secure web back. I'd blame the users (both corporations and personal ones) who won't upgrade, either to modern browser or a modern OS.
edit: Additionally those nice graphs are misleading. I bet other (SNI supporting) browsers have significant market share on XP too. Compare the market share of IE <9 (<20%) to XP (33%). And some of those IE8 users probably are on Vista/7.
edit2: And if we look at IE8 market share trend, it has been steadily decreasing for the past year. I feel safe to say that in a year it has decreased to a low enough percentage so that it can be ignored.
Pretty much sod all.
That's why users haven't forked out money. If it comes free with the new computers, fine. I reckon we'll be seeing it for another couple of years.
But don't blame the users, blame the utter lack of progress by MS.
SNI?
I think the UI improvements alone (desktop search, Aero Snap) are worth the price. Isn't this how Apple justifies charging for new versions of OS X? I still use XP from time to time and 7 is far and away a better experience. FWIW, I think it also provides a better experience than OS X, although, sadly, without the benefit of a POSIX layer.
"blame the utter lack of progress by MS"
It's not like MS has been sitting on their hands of late. Kinect demonstrates this. Windows 8 demonstrates this. I just don't think there's an appetite for MS progress or innovation. What people like about Windows computers isn't (and hasn't been for a long time) Windows. It's pretty hard to innovate when no one wants what you're selling.
To be fair, I don't "blame the users" for this. When MS had a tight grip on computing, they did little to make their products more compelling and they are realizing the consequences of this now: Windows' name is mud.
Good point. There is another side in here too: many people want to keep using the old systems they have learnt, instead of new better systems. Eg the backslash of Ribbon. It's pretty hard to innovate when on one wants anything to change.
I think that if something is better, then people will want it, steal it even. Take those un-skippable previews on some DVDs: people cite them as reasons to scrounge up pirate versions of DVDs. Oh, and how about those abominable "region codes" on DVDs? Are those "innovation" or merely a monopoly-enabled form of price fixing?
It's not clear, at least to me, that nobody wants Microsoft's innovations. I think it's more that they don't want them from Microsoft. Windows is a tough beast to change. The near ubiquity of the platform in the 90s and 00s means that there's a lot of people out there who "imprinted" on the Windows UI model.
It doesn't just affect Microsoft, though it may affect them disproportionately because of brand association. While OS X is a screaming success now, Apple's transition from OS 9 was anything but. It took 3 major revisions of the software before it was usable and a lot of people hated it at the time.
In contrast, Apple's introduction of iOS received a much warmer reception because there was no incumbent. Changing a major technology platform is hard.
To be fair, I think this is true across platforms. On what platform is window management better than it was 5 years ago? Sure, Windows lacks a good way to get an overview of all open windows (such as Expose), but I honestly think Aero Snap is a killer feature for desktop use (especially because it supports the keyboard so well). When programming, I often like to have my editor and reference material side by side and Aero Snap make this literally a snap, which is great when the reference material changes frequently. Sadly, when programming, I also like to use a suite of GNU tools, which makes Windows a bit less compelling.
Although you seem to think it's gotten less good, I still think keyboard navigation remains Windows' strongest competitive advantage over OS X. It's also the only thing Windows does better than Gnome Shell which is otherwise the best desktop UI I've experienced.
I primarily dislike the keyboard navigation in the 'Open File...'-dialogs. Tab is not as effective anymore since it takes more hits to change focus from the folder view to the sidebar. Furthermore, and this is nitpicking, the number of tabs needed also seems to be context depended and the number of items in the sidebar has grown.
I personally prefer StumpWM with KDE tools. I've never had a more effective keyboard experience.
And that's MS's problem, they've priced themselves out of their own market. The prices they charge are ridiculous and a little sad and pathetic that they think an OS, the thing that basically opens other programs, is really worth that much. I know that in reality it's much cheaper and provides more value than that. The expensive sticker price is the muppet's price, but that's the problem.
It's been coming free with our corporate PCs for years now, and we're still unlikely to switch any time soon.
(At least until they stop shipping security patches)
It's very disruptive, and horribly expensive.
If at all possible, you don't do it more than once per decade.
If you don't need to be on the cutting edge then there's a benefit to being a little behind the times and letting others iron out the bugs first.
I've been able to upgrade my Mac for about four years now from Leopard to Snow Leopard to Lion without a hitch. If I were using XP I'd probably have to buy new hardware with every upgrade and that's what's stopping people. It's not the money but the inconvenience that stops people. If their XP is doing the bare minimum of what they need it for then they're not going to upgrade to Win7 even if they know about the new ways it makes their life more convenient and how much prettier it is. They don't want to transfer their old important files from one drive to another. That's scary for the average person no matter how easy Microsoft advertises it to be.
It feels like Microsoft makes decisions with a heavier focus on marketing than real technical improvements, innovations, or user experience. I mean they make a good product (it's not my cup of tea but I'm not a bigot about it) but they're still operating like they have no competition and it's really screwing with people like us who are trying to move things forward.
I had a Dual G5 Power Mac from 2006. Consequently, I have not been able to upgrade that machine past Leopard. That's new hardware that had all of a single year of OS upgrades.
It's been demonstrated that you can run Windows 7 on Windows XP hardware. I know that when I upgraded from XP to 7 on a several-years-old machine when Win7 first came out, it was notably faster. I had installed the free beta, just to see, and was so surprised at how much better it was, I pre-ordered an operating system for the first time in my life.
What version of OSX can you run on a Mac from 2001?
However, for majority of e-commerce sites, even 1% would mean a loss in sales that would be unacceptable to many business owners I know
Source: I work in web hosting -- e-commerce mostly.
Especially when you consider that there are currently two forces in move that clearly work against each other:
On one end you have cool stuff like SPDY and more and more pervasive proxies altering HTTP data on the fly which makes SSL more and more interesting.
On the other hand, we are running out of IPv4 addresses, making it harder and harder to use one address per virtual host.
So even though we are much more motivated to use SSL these days, the growing difficulty of acquiring IP addresses makes it harder and harder.
I can't tell you what's more likely: To get both XP and Android fixed or to get wide-spread IPv6 deployment.
I have a feeling though that it's probably more likely that we see XP dead and Android fixed than we are going to see the current bulk of XP users (Corporations with crazy firewalls and users of pirated XP in Asia) having IPv6 connectivity.
I would love to be proven wrong though as I see IPv4 scarcity to be a much bigger threat to a free internet than anything else.
It wasn't my intention to pick you out specifically. I aimed that comment to the whole community.
>I can't tell you what's more likely: To get both XP and Android fixed or to get wide-spread IPv6 deployment.
I think that latter is far more likely. While IPv6 adoption has been relatively slow so far, still there has been constantly some progression. So I feel that there is fair chance for wide-spread IPv6 eventually. On the other hand, I'm almost certain that MS will not add SNI to XP ever.
It has been a while since I checked (I did a fair bit of SNI-related research for the HTTPS support in https://pagekite.net/), but at the time Chrome and Firefox on XP also lacked SNI support on WinXP for [edit: what I assumed was] this reason.
I think Chrome has since bundled their own crypto code so recent versions of Chrome on XP do work, but I am not sure about Firefox. I didn't test Opera.
This was also an issue on Linux desktops relatively recently, for alternative browsers like wget and KDE's Konqueror.
Any large corporation will have switched away from it by then (their risk assessments will force that), so that's the point you can start using the new functionality.
XP was released in 2001.
[edit for comment] TLS 2.0 which included SNI was released in 2006. Windows Vista which included IE7 was released in 2006/2007. [/edit for comment]
Microsoft may be guilty of many sins, this is not one of them.
SSL as a protocol is fully documented and standardized, and has been for a long time.
What evidence do you have that implementation would have been trivial?
A significant minority of users are going to be on old operating systems that can't or won't be upgraded to support IPv6, and servers will have to make the choice about whether to exclude them or not.