Flash has been in decline since the first iPhone, but is still used to track people with unkillable cookies and to make obnoxious ads. Hopefully those days are now over.
I wish Google and Microsoft would follow suit. Google probably will resist the most due to the entrenched interests of DoubleClick and YouTube.
>I wish Google and Microsoft would follow suit. Google probably will resist the most due to the entrenched interests of DoubleClick and YouTube.
This has been a feature in Google Chrome for at least a year. It isn't on by default, but you can change the browser to block any plugins until you specifically allow them on the page.
Partially yes. The problem with Flash ads is that they are difficult to block (without blocking Flash wholesale), because we don't have access to the runtime. There's no reasonable way for the browser to deep-inspect a Flash embed and determine whether or not it's abusive or relevant. There's also no way for your browser to alter and reformat content - it's either shown precisely the way Flash wants it to be shown, or not at all.
HTML ads will be rendered by software the user actually owns and controls - the browser market is healthy, largely driven by open source, and there is choice in your "runtime". Making all ads render in HTML will be a tremendous win for the user.
As someone else mentioned also, Flash is one of the worst pieces of software known to man when it comes to bugs and exploits. Nary a day goes by without a serious 0-day Flash breach being discovered, and it is still notoriously crashy.
I want advertisers to continue to use Flash, precisely because Flash is trivial to block and is almost always advertising.
There are approximately two exceptions, Youtube and Vimeo, and they can be handled with any of several click-to-play Flash blockers.
Making all ads render in HTML will make them much harder to block, because there will no longer be such a common telltale for animating, distracting cruft.
> scope (number of videos' that will play using it)
Many videos don't play in HTML5 on Youtube because of mandatory ads (displayed as overlays) and various settings available to right holders. The Youtube HTML5 player is perfectly potent...
> and style
... and on par with its Flash-based sibling. Now if you want to argue the Youtube player (be it HTML5 or Flash) is not as stylish and functional as Vimeo's, so be it.
I don't know for the GP bit I'd actually like the ads to be HTML instead. That would be the real end to the pages forcing Flash down your throat taking the content in hostage until you've seen the bloody 5sec ad. They'll find a way to force something down our throat anyway, avoiding HTML lessens the pain.
In my opinion a flash animation still outperforms a CSS-based animation in many cases. The problem with flash was never that it was inherently slow, but that people abused it to do slow things. On the security front though, you are absolutely right.
The performance side has nothing to do with animation performance and everything to do with the resources used to achieve it. Flash will murder a macbook's battery and generate an absurd amount of heat, doing nothing more than streaming h264.
That's so true, and as more and more ads go HTML you will see crappy devs animating the dom resulting in poor performance on iOS devices and damaging battery life.
So much of the advertising industry is so entrenched in Flash. It's a big beast to turn, especially when you consider that not just the last mile delivery servers need to support CORS and videos need to be in a compatible format (webm, please), but also the entire production pipeline needs to evolve their processes and tooling.
I've been kicking and screaming to get more of our partners onto HTML5 for a couple of years now. I've met with some success, but there's an awful lot of hoop-jumping to get there. Many advertisers are still of the mindset that "HTML5 is only for mobile", if they're even aware of it.
Maybe once this battle is won we can look at a smarter VAST standard!
Flash has been in decline since the first iPhone, but is still used to track people with unkillable cookies and to make obnoxious ads. Hopefully those days are now over.
No, they’re not. HTML5 supports session storage which will lead to some very nasty cookies stored for an eternity on your machine.
Google bundles their own build of Flash in with Chrome, the normal Flash plugin doesn't work with Chrome. It's the only browser on Linux that still gets new (non-security updates) versions of Flash.
I wish this was working properly. I am using the HTML5 trial and embedded videos will randomly not work, and many, many other videos will simply show static with "You need Adobe Flash to view this video."
For some strange reason, the exact same videos work fine on my iOS devices — just not on a no-flash desktop web browser. I'm suspicious that YouTube purposely allows a larger selection of their content to work with iOS using HTML5, and purposely fails/requires flash on a desktop under HTML5.
I get the opposite, I get videos that play fine for a web-browser, but won't work at all on mobile, even though it should. I've even taken a video, uploaded to my account and it STILL says "the owner has indicated that this video not show on mobile" or words to that effect. Though when I uploaded the video, there was no setting to click to make it unavailable to mobile users.
Nice to see proper flash blocking in Safari. Previously, I ended up deleting the plugin and having to wipe out the package receipts to convince Apple to stop re-installing the plugin.
The strategy of compressing RAM pages before resorting to swapping them out is a nice addition (discussed on p. 17 of the review). Something similar is in the works for Linux as well: http://lwn.net/Articles/545244/
The other main highlights from my perspective: "App Nap" energy-saving API (p. 13), generally better battery life, even on old hardware (p. 18), & support for offline speech-to-text (p. 23).
These kinds of things are best handled as an integrated component of the operating system, rather than a hacky third party add on, and are likely more viable now that we have so much more available CPU time, on multiple cores so it doesn't even impact latency, and better compression algorithms.
Also note that Ram Doubler, per the Ambrosia Software description, didn't just do compression, it also re-used free RAM that other applications had allocated to them while they weren't running, depending on the cooperative multitasking nature of the operating system to avoid conflicts. Before yielding, it would either compress or swap out that "borrowed" RAM.
A properly virtual memory system and cooperative multitasking make these techniques not particularly useful, while improvements in technology since have made compressed RAM more useful.
I wonder when airplanes are going to be new again, or books, or cars. Maybe you mean everything related to computers? I am not sure of even that though...
Concorde was a good try. It failed for political and sociological reasons, but hypersonic airplanes were set to revolutionize air travel until it became politically unpopular to support. We're facing down a generation designed around space-based air travel, though, so it could happen again.
Which is frankly normal and good. I wouldn't be surprised that Apple has been playing with this type of feature (as a module) for this amount of time as well. In Linux, a feature doesn't (normally) get promoted into the kernel until it's had some time to soak in user space for awhile.
Yes, one of the reasons I ended going Windows 7 instead of Apple was that they only supported OpenGL 2.1 for a very long time.
Apple systems still leave a lot to be desired if you are serious about 3D work. And the new Mac Pro does not seem to be worthwhile of a 3D workstation.
Yeah his reviews are very good. I do feel it would be nice to have a bit of a comparison section at the end though, to discuss some of the more technical details within the operating system landscape. For example timer coalescing/race to sleep is something I know Windows 8 has and I assume Linux, and the App Nap sounds like a more advanced version of the Metro tombstoning, and I think Linux can do memory compression but Sprite Kit sounds rather unrivalled by first parties. I would love to have his views on it all.
Seriously? Thank your lucky stars you were born in the correct generation. You used to have to read a page of text just to get advertised to. http://imgur.com/ljQZgEJ
For the last OS I loaded it up on my iPad and then took it with me for a weekend, reading it whenever I had downtime (we were doing a lot of shopping so there was a lot of downtime).
Managed to finish it up by Sunday. I'll probably do the same this time.
Easy install on my ML 2011 iMac i5 6970GPU system.
Initial impression, scrolling certainly feels better. Do wish they would leave my desktop background to what I had it set to. Other than that, being an OS upgrade where I don't notice anything odd is key to my satisfaction.
5+ gig download, took about forty minutes, somewhat less I think as I walked off, to install.
That's strange - my desktop background wasn't changed by the install. Perhaps you were using a ML-included desktop image that was removed in Mavericks?
After downloading, it took roughly 20 mins to install.
> In the years that have passed since then, the Mac has
> indeed been on a steady march toward the functional ideal
> embodied by the iPad, a product that is arguably the
> culmination of Jobs' original vision of personal computing
concerns me quite a bit. We all know Jobs' original vision of personal computing was a tightly locked-down walled garden, and I can't help but think that inching towards this destination is inevitably a change for the worse.
Think about it. With the drop of the non-Retina display MacBook Pros today, no Macs are now officially user-upgradeable.
What was the reason given?
2mm in thickness. Two. Fucking. Millimeters, so you can stare at the edge of your laptop with a hard-on. Oh, and that absurdly high resolution display that you'll need a goddamned loupe to appreciate.
All kernel extensions now must be signed in Mavericks. OS features brought about in Lion still bug me, like the absolutely back-asswards autosave system that uses duplication, and the lack of direct manipulation while scrolling. Also, Gatekeeper is a huge uh-oh.
It's the reason why my MacBook is now sitting in a closet, and why I'm using a 2005 Toshiba Tecra with Debian on it. Amazing how Linux news has gotten so rare these days... but stick an Apple sticker on something and it shoots to the top of the front page. Sad.
Apple TV with Airplay in theory is awesome for presentations. I've setup Apple TV on a bunch of LED screens for a client, with the intention of using Airplay for displaying presentations from Macbooks.
So far, the solution has been close to useless. The lag is unbearable, even for simply actions like changing webpages and PPT pages. In the end, users end up going back to HDMI or even Thunderbolt to DVI/VGA connectors for presentations.
To make it worse, it seems that different versions of OS X will give a variety of glitchy behavior on Airplay.
This probably has something to do with your network setup. I've been able to get the lag down sufficiently that some games are playable, though I'd never recommend playing a twitch-based FPS over Airplay.
For me what really made a difference was to ensure that only the laptop was connected to my network via wifi. My Apple TV is connected using ethernet. Before I made that change, I experienced significant lag, but now the lag is perceptible but pretty low.
AirParrot has done this for a while. Nice additional feature not included in Mavericks: you can send just one specific application window to the appletv.
If it's anything like the current offering, mirroring via AirPlay to an Apple TV, it will be useless. There's a very noticeable lag. OK for presentations but as a a second monitor it would be dysfunctional unless they put in a lot of latency work. I hope they have.
I always love the use of the word "useless" on this site. "Ok for presentations" means it has an obvious use: presentations. How can that be useless? I'd agree it can be better, but it's obviously not without a use.
I specifically said useless for dual-screen setups in response to the comment I was replying to. I too find un-analytical praise or criticism annoying. I hope I wasn't doing it.
Just tried using airplay as a second display on my prev-gen Apple TV (black puck but 720p) and new MBA. Works fine but noticeable lag that makes mousing nasty. You probably don't want to use this for a "real" interactive display. Maybe the newer ATV would be faster, but I wouldn't expect too much.
I thought readers might appreciate a report from an actual end-to-end test. There's a huge usability difference between 15msec and 150msec latency. I observed the wrong end of that range.
FWIW, I think video stream buffering is the culprit rather than inherent channel latency.
Depending on the USB input device, that's not necessarily true at all. You could very easily find crappy USB input devices with greater latency than your ping to the news.ycombinator.com server, unless you're in Antarctica or something. There's at least one John Carmack twitter rant about ping to Europe being better than latency from some USB input device.
If you're a single router hop away from something the latency can be imperceptible. Whatever trouble is involved here isn't down to network latency.
We have Mac Minis attached to our TVs. And infuriatingly, you can't use OS X as an AirPlay target, only as a source. I'd love to see that change, but don't see any mention of it in Mavericks.
Yeah, we may well end up using it, but it just seems like an annoyance to pay for something like that.
Also, we have an issue with the TV computers not being on the same subnet as wifi'd laptops, so they don't appear through Bonjour. But that's a whole other issue for enterprise that Apple doesn't really touch.
One benefit AirServer has over any of Apple's AirPlay implementations to date is that it supports 1080p rather than just 720p like the Apple TV when receiving data from an iPad or iPhone. I specifically have an Apple TV + AirServer hooked up to a wall-mounted screen instead of an Apple TV just for this reason.
For making such a big deal about resolving the multiple desktop and full-screen issues, Mavericks feels a little disappointing. Switching between full-screen windows is still accompanied by the painfully slow animation which still can't be disabled.
Trackpad scroll speed on my 13" MBA is also noticeably slower without significant load on the machine. This seems deliberate, but it's a move in the wrong direction for people that already have the trackpad sensitivity maxed out.
If that were the case, it would be a matter of overly-aggressive optimization on Apple's part. Full-screening in the days of multiple monitor setups shouldn't necessarily mean that windowed apps get paged out (which is about the only reason I can see for causing a slower transition).
Check out TotalSpaces which fixes a lot of the 'problems' with spaces introduced in Mountain Lion. In ML at least (haven't tried v2 on Mavericks yet) you can completely disable the transition animations when changing spaces.
"If you prefer the old behavior, uncheck the "Displays have separate Spaces" checkbox in the Mission Control preference pane. Doing so will also restore the ability to have windows that span more than one display."
Its interesting how Scott Forstall has become the goto name to sully in many of these articles. He pretty much pioneered iOS but one redesign later he's nothing in the shadow of Ive.
The only mention of Forstall in the review is on the first page.
> But that was all before last year's ouster of Scott Forstall, senior vice president of iOS Software. By all accounts, Forstall was one of the driving forces behind the iOS aesthetic that Lion and Mountain Lion so enthusiastically embraced. Jony Ive's iOS 7 strikes off in a bold new direction based on a philosophy that Apple is eager to generalize to the company as a whole—leaving OS X holding the stitched-leather bag.
Not exactly sullying his name. There has been a design shift, it's fairly striking, at least on the surface, and it seems to be tightly coupled with a changing of the guard.
The story that will go down in history is Scott Forstall was the skeu guy, and he got the boot so Ive could flat-iron everything. We don't know if it's true, but unless Apple tells us otherwise, it fits the data we have better than most things.
Actually, Forstall will most likely go down as leaving because he was power-hungry and didn't play well with others. Skeu probably had little to do with it (imho).
Way way off topic, but: Can we cut the IMHO shit? Why do we have to qualify every opinionated statement as opinionated? If some mouth breather doesn't understand that what you wrote is your own opinion that's on them not you.
I think it's just used to mitigate the edge of an assertion. Or make it seem more like a declaration than some point you're willing to vehemently defend.
In other words, imho I don't think it really means
This is something I despise about our culture. I see this everywhere and not just with imho. People tack on qualifiers all over the place in order to give a hint to the reader that they are expressing an opinion. I wish people would just give each other a break and that people would not be so insecure about appearing to be wrong in an argument.
People make baseless assertions and throw around absolutes all the time. It's not the end of the world when an absolute is inaccurately applied to a relative quantity or quality!
I think an IHMO is far better than the kind of crap creationists come out with. Its usually opinion they are regurgitating, and they talk of it as if it is fact.
If one isn't strongly opinionated, why are they weighing in? The way you describe it, it seems like a cop out: "here's my baseless opinion that I'm using to drum up really important internet points, but don't skewer me! I'm just a humble baby!"
It's that time again. I'm glad they do paid eBook versions now as I'm not an ars subscriber, but they definitely deserve some money for putting the OS X review together every release. I love reading Siracusa's minor gripes and grumbles, and when he feels something deserves genuine praise.
> "Some people think Ars Technica forces me to break my article up into many tiny pages. That’s not the case. I choose how to paginate the article. I like to break it up on logical section boundaries, which means that the “pages” vary widely in length. I do try to keep any single “page” from being too short, however."
I've only been using 10.9 for a few hours but I have already noticed that some of the unnecessary decoration was removed. Notepad, the notification sidebar, launcher, widgets screen, etc have been toned down.
Mavericks GM hasn't been too good for my 2011 MacBook Pro. This machine has only 4 GB RAM, and it shows. It's swapping noticeably more than before, and overall everything feels less snappy than on Mountain Lion.
On the other hand, the battery life is definitely better. It's not really worth the performance hit, though...
Late 2010 MBA using the old Mavericks GM - 13A598 - is fine. No different than 10.8 at all. 4 GB RAM on a C2D. I really don't think I would have even noticed the difference had I not had to run the installer.
Interesting, the RAM compression should actually increase performance. Well I would highly recommend an SSD for you, with 4gb RAM my Macbook Air flies, and it's all thanks to the SSD.
I second the SSD upgrade. I regularly use three different MacBook Pros -- two with standard hard drives and one with an SSD. The SSD makes a world of difference. It feels like a new machine. I never have to wait for anything, not even Photoshop or Word. With the standard hard drives I always get spinning pinwheels, 5 minute boots, etc. In my opinion, even the 2013 MBPs are unbearably slow with standard hard drives.
I got a couple of 2TB WD Greens for $85 apiece and a ZyXEL NSA221 two-bay NAS (some ARM chip running what I think is Debian but I've never bothered to look) for $120. I think that one's discontinued now, though.
You need to define what you mean by "big enough" and "affordable", considering there are now 1TB SSD for ~$500, which I personally consider both "big enough" and "affordable enough"
That's plenty big enough, but kind of pricey compared to $65 for a 1TB hard drive. Try to get either of those capacities in a laptop (I take pictures when I travel), and the gap gets even bigger.
Any incompatibilities people have run across yet. I want to figure out if there are any obvious dealbreaking changes before I spend time making a backup and upgrading.
pkgsrc also works, supposedly. I'm currently using Macports on 10.6, but wondering whether I should stay on macports (known quantity, huge number of ports), switch to homebrew (hype, easy to write & push new packages, lower number of ports) or switch to pkgsrc/pkgin (cross-platform, good number of ports, but very low popularity & likely support)
If you're going to give us a 24 page whitepaper on a free OS upgrade, at least give a brief intro about your findings... verbiage and metaphor excluded.
Herp, derp. Siracusa has been somewhere between critical and brutal toward the Finder (desktop slash file browser) since Mac OS X was born. He is also very well-known for being meticulous. Your comment reflects poorly on you.
tl;dr: It's great. Not much has changed in a major way, but there are improvements all around. The performance and battery life of the OS is much improved. It's free - and you should upgrade.
The line between "rambling" and "detailed" is just how interesting you find the subject. If you're not interested in reading a 24 page review, you should probably not read a 24 page review.
Python 2.7.5 (default, Aug 25 2013, 00:04:04)
[GCC 4.2.1 Compatible Apple LLVM 5.0 (clang-500.0.68)] on darwin
Type "help", "copyright", "credits" or "license" for more information.
>>> a = 1000
>>> a/1000
1
>>>
Yeah, I realized this after posting. Mavericks is installing its own version of python compiled with clang-5.0. My path was modified by the python installer (from mountain lion) to point under Library and was compiled using the older gcc 4.2. Looks like that program crashes for simple things (division, addition etc.).
There is a problem with the readline emulation of libedit that causes python 2.7 - 3.3 from python.org to crash under 10.9
http://bugs.python.org/issue18458
I get Segmentation fault: 11 if I enter more than one line in python3.3 running interactively in Terminal, but ipython3 seems to work.
the opening 5 paragraphs into this article was infuriating -- from cats having 9 lives, to self-actualization, the after life -- get to the point already. it's an operating system. a new version is out. talk about it.
Au contraire: The opening five paragraphs set the stage for everything that follows by providing context. This is eloquent and lively writing from a master practitioner.
The author's been reviewing OS X like this since 10.0. Probably involved with Apple's OSes since even further back. It's clearly become something dearer to him than a child.
Anyone had random logouts when using Expose?
Not sure if it's USB DisplayLink adapter, or some weird bug in OS X.
It's quite rare, perhaps twice or once a day. Haven't lost important data yet, but I feel it's coming.
Interestingly, he doesn't type all that. He dictates everything. So essentially, he's talking for 24 pages!
I always look forward to Siracusa's OS X reviews. It's fun to read his past OS X reviews, especially of Tiger, Leopard or even all the way back to Panther and Jaguar when OS X was still in its early days.
He goes to some pain to make the page breaks logical and helpful rather than click whoring. He has discussed this on podcasts at length, most recently on ATP. I appreciate this may address a point you aren't making.
The 2nd monitor upgrades and fixes really help out a lot. I think as a developer who works mostly with 2 monitors now working in full screen mode both monitors can be properly used.
" Frankly, this entire window is a user-interface disaster. And we haven't even mentioned the checkbox to the right of each label. Can you guess what those do? (No, there's no tooltip when you hover over one.) I'll spoil the surprise. When that box is checked, it means the Tag appears in the Finder sidebar; unchecked means it doesn't."
I think he overlooked the text right on top of the menu, which says "Show these tags in the sidebar:". Pretty obvious to me.
Know it's too early, but can anyone comment on Screen Sharing improvements in Mavericks? I regularly access a headless mini and have had to occasionally kill screensharingd for hanging sessions, and/or lose connectivity on occasion for whatever reason. Screen Sharing's been improved with every OS X release, but it's not spectacular.
John had had a lot of complaints regarding how his Kindle version of a previous review was not available for the iPad [1]. I wonder if this is still a case.
EDIT: Seems to have worked fine - download in progress, and I didn't actually complete installation of the free app (AFAIK; I did have to enter tombstone info, but I can live with that).
I've a S2 and an Asus 300T: I read ebooks via Google books on those devices.
Music is streamed through rdio, which allows me offline access to several GB worth on my phone.
I have Netflix.
I do not own any iOS devices.
I see no reason why I would ever want to buy anything through iTunes, since I am happy with my music service, with my books account, and with OSX and the few apps I've added to it.
Yeah, I know, I am a 1% outlier (Android Google mobile user, OSX only for the reliable hardware for business and not for personal use). But why use iTunes to distribute Mavericks when the App Store is already there?
I'm not sure that I understand the relevance - or the practicality - of that.
One of the reasons I finally dumped my HP PC and Ubuntu was that I no longer wanted to play sysadmin. I wanted my computer to "just work" with minimal (non invasive) effort on my part.
So far, I am supremely happy. This Air is the bomb: Fast, silent, robust, well made, wow. Nice piece of work.
But needing a CC to get a free OS update is neither non-invasive nor effortless, which the platform had been up until now.
(Sure, I could get the update from somewhere else, but I would only be doing so because Apple failed to address my particular use case. I'd rather get it from them, jic - as in "just in case something borks in the update" - then I can go back to them with ground upon which to stand. Sure, as noted elsewhere, my use case is in their 1% outlier fringe, but that doesn't mean I wish to be without support or without minimal privacy invasion.)
You know, you're right but until I saw parent's comment I didn't realize why I found it hard to scan through the first page of the review. I wonder if orange is harder to scan through than a blue link with black text.
And sometimes it does seem there are links to excess.
Boy that first page was completely worthless. Maybe there was one useful sentence in there, saying there are new features and bundled apps. Not sure I should bother reading page 2. Ars' latest iPad announcement coverage was awesome, though.
He’s not joking. Later in the first page, Siracusa says
‘Mavericks is the first California-themed release of OS X, named after "places that inspire us here in California," according to Craig Federighi, who says this naming scheme is intended to last for at least the next 10 years.’
And the words “California surfing spot” in your quote link to http://en.wikipedia.org/wiki/Mavericks_(location), so that is the name of the place. I guess it’s just coincidence that “maverick” is also a word.
http://cdn.arstechnica.net/wp-content/uploads/2013/10/safari...
http://cdn.arstechnica.net/wp-content/uploads/2013/10/safari...
Flash has been in decline since the first iPhone, but is still used to track people with unkillable cookies and to make obnoxious ads. Hopefully those days are now over.
I wish Google and Microsoft would follow suit. Google probably will resist the most due to the entrenched interests of DoubleClick and YouTube.