Download this Episode
You can also download this episode as a standalone MP3 file. Here’s the link:
SitePoint Podcast #63: There Are Two Webs(MP3, 54.3MB, 59:15)
Here are the topics covered in this episode:
- Google Introduces WebM Video Format
- Google Introduces Google Font Directory
- First Look at Google Chrome Web Apps
- Twitter API Basic Authorization Ends in 5 Weeks
Browse the full list of links referenced in the show at http://delicious.com/sitepointpodcast/63.
- Patrick: 99designs signing “We are the Champions”
- Stephan: HTML5 demos and examples
- Brad: Face Recognition API
- Kevin: Video JS
Kevin: May 28th, 2010. The Google IO conference brings big news for web video, fonts, and applications. I’m Kevin Yank and this is the SitePoint Podcast #63: There Are Two Webs.
And welcome back to the SitePoint podcast. As always I am Kevin, and I’m with Patrick, Stephan, and Brad once again. Hi guys!
Stephan: Howdy doody.
Patrick: Hi Kevin.
Kevin: I missed you last week.
Patrick: Did you?
Brad: You can be honest.
Patrick: Had a nice vacation? No.
Kevin: Last week was your live cast from WordCamp Raleigh. How did that go?
Patrick: It went great actually. It went really smoothly. We had a lot of help from people on the ground, let’s say, that know the technology of podcasting far better than we do.
Patrick: Like Dave Moyer from Bitwire Media, and an AV assistant that was on site, and Dave basically ran the technology for the show; mics, mixers, USTREAM video, and so on. We focused on the content. We had 11 guests and gave out 33 prizes and it went surprisingly smooth I’d say. You can watch a live — you can watch the broadcast recorded on USTREAM, it’s at ustream.tv/recorded/7124319, or if you go to sitepoint.com/podcast and scroll down a little bit you can see the live stream and you’ll be able to find it from there.
Kevin: Hmm-mm, yeah. I heard the audio sounded really great. Those guys who were helping you out with the set-up did a really good job, really impressive work.
So that live podcast is unofficially dubbed podcast #62. So if you’re seeing this one, 63, what happened to 62? 62 was that live podcast. But if you don’t feel like watching the entire two hours on the USTREAM website, rest assured the highlights of that show will be released in future interview weeks, right Patrick?
Patrick: Yeah, I think we’ve decided to release them in probably four episodes, the interviews themselves in chunks about — basically broken up into segments of interviewees; so the WordCamp Raleigh organizers, book authors, podcasters, theme guys, and maybe video bloggers. So it might be four to five grouped together by subject, but they will be out in the next probably two months I would guess.
Kevin: Hmm-mm. and how was the rest of the conference? Did you get to take it in at all or was it all live broadcast mania?
Patrick: Well, it was a lot of fun. It was good to meet Brad for the first time in person. I’ve met Stephan before at conferences a couple of times, but this is the first time we got a chance to hang out, and we all, me and Brad both had talks, I think they both went well and I enjoyed Brad’s and learned some things, and yeah, I go for the networking, to meet people and talk and it was a lot of fun.
Kevin: Brad you must’ve been in your element at the WordPress conference.
Brad: I am definitely in my element. I would go to one every weekend if I could to be honest. They really are a great time, and I mean like Patrick said, not only the presentations but the networking really is invaluable. So if you’ve never been to one I highly recommend finding one around you; there’s one or two almost every weekend, so at some point one will be nearby so definitely be there.
Kevin: But there was another conference going on last week, and if the organizers of WordCamp Raleigh will excuse me saying so, I think it was a bit bigger. The Google IO conference was stealing headlines left and right. I think when Google decides to do something like that just all eyes turn towards them inevitably because I didn’t hear anything that didn’t originate at the Google IO conference last week I don’t think. We have picked and choosed the highlights, the things that stood out for us for web developers from that conference.
And the first one was Google’s new web video format. Now, Stephan this is something they had predicted was gonna happen, right?
Stephan: Yeah, I think so. I think that people were kind of expecting this and it uses Vorbis, right, for the audio.
So, I mean I think people kind of thought this was coming, and I guess it’s kind of good to see, you know.
Kevin: As a recap, all of the browsers had settled on H.264 which is a commercial, patented video format that has been licensed royalty free to content producers on the Web, so we are free to host H.264 video and produce it as much as we like for free at least for the next, I think, till 2015 or something like that; for quite a while anyway. And just the browsers that embed support for playing this back need to pay a licensing fee to the owners of that format. And this was not satisfactory to people who are, you know, very standards minded and want the Web to be free and freely implementable. The theory is that you should be able to write your own web browser without having to pay anyone for the pleasure of using that technology.
And so browsers like Firefox had put their foot down and said, look, we’re not going to support H.264 video, and the only alternative for a while there was Ogg Theora, excuse me, it’s difficult to say; Ogg Theora (laughs) which while free at least for the time being, and there’s some question as to whether any high compression streaming video format can be entirely free of patent encumbrance, but at least for the time being it was free, but the consensus was that it kind of sucked too.
Brad: Yeah, I think this is pretty exciting. I mean video has been one of the biggest question marks with HTML5 for a while now, so hopefully there’s a clear defined winner, and it looks like just about everybody’s supporting them, so it’s really exciting for HTML5.
Kevin: Yeah, everyone except Apple. Apple is the big holdout this time around. So, just to lay it out there, WebM, this new video format that Google is introducing is actually sort of three technologies bundled together. You’ve got, as Stephan mentioned, the Ogg Vorbis audio codec, so the audio portion of the video that you’re hearing is encoded in Ogg Vorbis rather than say MP3 or something like that. So, I don’t know, if the fans of Ogg Theora needed some consolation there’s still some Ogg in this, (laughs) where Ogg Theora was the video format, Ogg Vorbis is the audio format. So Ogg Vorbis is considered pretty decent and that made it in.
The video format, though, is VP8 which is a video encoding technology that Google bought and that everyone was expecting or suggesting at least that Google should open source for the good of the Web, and that sounds — that seems to be exactly what they’ve done.
And finally there is the actual file format that bundles together this audio and this video and that’s called the container format. And it look like they have reused and enhanced the Matroska file format which is also open source and quite popular. But it was always kind of second fiddle to these MPEG-4 formats that are used for H.264. So is — the consensus, and I’ve linked in the show notes to a really big analysis of this WebM video format and how it holds up against H.264 from one of the developers of x264, which is an open source software package for creating H.264 video. So, dedicated as he is to the H.264 format, he sat down and read through the source code of the — at least the VP8 video format that’s going into WebM. And he had a few choice words. In fact, he had a few tens of thousands choice words reading this thing. It’s a lot to get through, but if you skip to sort of the summary sections he seems to grudgingly admit that the technology is actually pretty good here, but the code, the source code, is kind of messy. So if you were going to try and implement your own VP8 video player or video encoder, you had best plan for some late nights reading ugly C source code.
Stephan: Does he expect the source code to improve? I mean I would assume it’s going to over time?
Kevin: Well, it seems like reading it, or skimming it, if I am to be honest, he went in expecting that VP8 was actually quite an immature format and that it had a lot of room to grow. But he was surprised to find, first of all, that a lot of the obvious things you can do in a video format to save space and improve quality have already been done. So he says he expects I think they’re about 80 percent of the way there as far as taking advantage of commonly known video compression techniques to get quality and compression. So he was surprised that they were that far along. He said there’s still plenty of room to go with some really advanced techniques if they are able to implement those in a patent-free, open source way, and that’s what formats like H.264 are sort of taking on at this point. So if they go down that road they can make similar great advances coming forward.
But the other thing was that he thought that this code would be really immature and just sort of thrown together in the last couple of years in a bit of a hurry. But he found there were comments in the source code dating back to, oh geez, 2004. The source code was a lot older, yeah, as far back as early 2004, he said, this source code was a lot older than he was expecting and that made him sort of take a step back and go, whoa, I can’t pardon this code for immaturity; it should be pretty mature by this point. And yet there were certainly messy signs of hackery that he was hoping he wouldn’t find in a mature code base.
So I don’t know what we can expect from a free format, you know, beggars can’t be choosers, right?
Stephan: Yeah, but I mean it’s an alternative, right? I mean it’s not H.264 and it’s something that we can all use and I mean I see it as a good thing.
I don’t know, I mean at the same time I kind of see what happened with Ogg Vorbis too, you know, it kind of died there for a while and no one used it because MP3s became so prevalent. So I’m torn, and I think that Apple not jumping on this and trying to maybe push, since they’re trying to push HTML5 so much, why they’re not pushing this as another open technology I don’t know. So maybe they want to put their boot heel on the throat of Google for producing all these phones (laughs).
Kevin: Reading about it Google announced this and in tandem Firefox announced they’re gonna support it, Opera announced they’re gonna support it, even Internet Explorer said that if you install the VP8 software package that teaches Windows how to play this kind of video, Internet Explorer 9 will support it. And just Google VP Linus Upson says “it’s not a technical challenge; if you look at the other browsers they’ve already implemented VP8, it’s just been a matter of a few weeks.” And I think in the week following Google IO someone said to — well, someone posted one of these Steve Jobs emails where they emailed Steve Jobs at his personal email address and said “Hey, Steve, why aren’t you supporting WebM?” And reportedly he replied just with a link to this analysis of the VP8 format and how messy the code is and all of that.
So it sounds like Apple is kind of holding its breath. I would be willing to bet that if the content producers play along and start converting all of the content from H.264 to VP8 that we might well see a Safari release. But for now Apple is hedging, it’s keeping its bets. And I can’t blame them because all of their mobile devices, all of the iPhones, all of the iPads come with little chips in them that play H.264 really quickly on minimal battery power.
Stephan: Well, yeah. When you invest that much money on something of course you’re gonna want to keep that technology around, right? So it’s obvious why they’re not supporting this.
Brad: But I mean how long can you hold out if basically every other browser out there is going to support this, how long can Apple hold out and say no we’re not? Eventually I think just the pressure is going to get to them and they’re going to have to.
Patrick: You can probably hold out until people like developers and people who make the content and release it actually start to use it in a wide way. I think that’s when it’ll become an issue. Then again, most sites use Flash and Apple doesn’t seem overly concerned with that, so I don’t know.
Stephan: Can they do H.264 you think on the iPads or whatever and then release something for Safari so that it at least supports the format? I mean that doesn’t seem like it’s too difficult just to support the format.
Kevin: It seems like they could if they had to, but it would kill battery life because those devices don’t have hardware decoders for this new VP8 video format. I’m not even sure there’s such a thing as a hardware decoder for VP8. And on today’s Web does it make sense to be standardizing around of format that doesn’t have low-power hardware video decoding on mobile devices. I mean obviously that will come; that will come if this is a successful format, but right now it’s a desktop format.
Stephan: But I mean why can’t they just release a Safari plug-in or an update to Safari that at least supports it in the browser on a Mac or even on Windows; why can’t they do that?
Kevin: They can, but do they want to? Because as we’ve said, obviously if this format gets wide acceptance they are going to have to redesign all of their mobile devices to have new chips in them for playing back this new video format. So it seems like they’re hoping this format will stay the second choice, and so people will continue to encode in H.264 first and then now use VP8 instead of Ogg Theora to support the other browsers like Firefox and Opera that require it. So it’s still a pain for content producers to have to produce two formats for every one of their videos.
And Apple seems to be hoping for that (laughs), which is a shame. It’s a shame.
I don’t know, is there something special about video? Apple seems to be of the opinion that video is special, that we need open, patent-free formats for HTML, for images; all that is fine. But right now video is just too complicated and that it is sensible for everyone who builds a web browser or any device that contains a web browser to pay a license fee to some company that is responsible for that video technology. Is video special; does it get a pass?
Brad: I think it’s a much more complicated thing to do on the Web is work with video, so I think that’s always been kind of the issue is that it hasn’t been to where you have a majority of people that might want to help work on an open source project or something like it, because it is much more involved. So, you know, that might be a part of it.
Kevin: Well, I mean if you took the — if everyone on the Web took the same hard line that Mozilla and Opera are taking, that they will not support or endorse any video format that isn’t completely free, then we wouldn’t have iPhones that can play video. We wouldn’t have iPads that can play video for ten hours on one battery charge. These devices wouldn’t exist because the technology for doing low power hardware decoding of video is entirely locked behind patents; it’s owned by commercial companies right now, there is no open source format that let’s us do that. So video on mobile devices would be where it was five years ago right now which is basically nonexistent. And would that be acceptable? Because that seems to be what Mozilla and Opera are saying we need to do for the good of the Web.
Stephan: Well, we’ll just wait till Google releases a chip that on an Android phone plays this new format.
Kevin: That would be the big game changer.
Stephan: Yeah. I mean and then you’re going to have people clamoring —
Brad: Checkmate Apple.
Stephan: Yeah, checkmate.
Patrick: iPhone market share drops from 97 percent to 95.
Kevin: But I have to congratulate Apple for pushing the game forward. They sort of say look we’re not afraid to embrace a little commercial technology to show where these things can go and prove that video in your pocket is something that people want, and then they allow the standards bodies and the Firefox’s and Opera’s of the world to bring up the rear and insist on free and open technology for doing this stuff before they support it. But meanwhile Apple is pushing on to the next thing; it’s like there are two webs, there’s the proprietary experimental web that Apple is playing with, and then there’s the open standard web that Firefox and Opera are playing with, and as a user you’re kind of expected to choose whether you realize what you’re choosing or not.
Patrick: What type of web are we playing with?
Kevin: Yeah. Well, as content producers we’re stuck in the middle. Right now we have to produce content for both formats in order to support as many browsers as possible. So it is the content producers who are suffering here. Oh, it is a tough life; it’s a tough life. And (laughs) Google is trying to make our lives easier with their Google Font Directory which is the second big thing that they announced at Google IO last week, and this one was really exciting to me.
Who else is really excited about the Google font directory? Am I the only font nerd here?
Brad: You love your fonts.
Kevin: I do! I do! I don’t know why I’m in podcasting, there’s no fonts in podcasting.
Patrick: It’s a catchy name, fontcast.
Stephan: Yeah, it’s a fontcast, yeah. Kevin’s font hour. Yeah, I mean this is cool and all, I just don’t know if I’m — the cool factor hasn’t hit me yet I guess. It looks neat but I gotta use it.
Kevin: So the idea here is you can go to code.google.com/webfonts, all one word, and there is a collection here of some 20 or so fonts that Google has decided to foot the bill for. We’ve seen in the past year services like Typekit launch and enable web designers to go and pay licensing fees in order to include these new custom commercial fonts in their web designs. So if there’s a particular font in the Typekit library that you really like and you want to use on a website that will display in modern browsers that support downloadable fonts, well, Typekit will let you pay them for that privilege, but it still — that’s a bit of a mental barrier, I think, when all browsers support four maybe five fonts that you can reasonably assume will be present on the system, people tend to limit their thinking and their design choices to those free fonts. And making the decision up front to go, alright, I’m gonna dive into that Typekit library and find something more suitable to my brand to put on my website, and I’m going to plan to pay for that, it’s a tricky decision to make up front, and I think that keeps a lot of designers from diving into these custom fonts.
But Google now is saying forget about all that paying stuff, we’re paying for these fonts. Not only are they paying the licensing fees on these fonts, they are also hosting these fonts for you. So all you have to do if you want to use any one, or any combination of these fonts in your web pages, is include in your webpage a link tag that links to a piece of CSS code that Google again hosts, and it includes that font in your page and then you can go nuts. So it’s like we’ve just gotten an upgrade and every modern web browser now supports 20 more free fonts.
Brad: Even IE 6.
Kevin: Not IE 6, I’m afraid, no.
Brad: No, it is supported.
Kevin: Oh yeah, you’re right!
Brad: It says on FAQ.
Kevin: I keep forgetting that Internet Explorer has supported downloaded fonts longer than most other browsers.
Stephan: So this is cool and all, but what is Google getting out of this? That’s what I — they gotta —
Patrick: Google loves us. They just want a better web for you and for me; your children Stephan.
Stephan: No, no, that’s — that is a dangerous idea, right; there’s a saying, right, that corporations are like bricks, bricks can’t love you back. You can love the brick but they’re not gonna love you back.
Kevin: (laughs) I haven’t heard that saying. I have heard don’t look a gift font in the mouth.
Patrick: Who coined that, Kevin Yank?
Stephan: So I’m just like what is Google wanting from this?
Kevin: To hear Google describe their strategy, anything that gets people to browse more web pages that contain more Google ads in them the better it is for Google. So everything they do is geared to make people more addicted to clicking the next link in their web experience. I don’t know if fonts do that.
I agree with you. This kind of stands out as a ‘huh’?
Stephan: This is a little strange for them. Almost as strange as Web TV; it’s a little weird.
Patrick: I think a lot of these fonts are accent fonts. I mean if you open up the previews when they get down to the 12 or the 14, kind of the normal text size, one might say, they’re just not as clear as some of the standard fonts I guess. I don’t know, I think it’s cool to have variety though, and it’s cool to have it work with everyone, so it’s a good thing, but like Stephan I always kind of look at Google and wonder.
Kevin: There are at least two fonts here that are really, um, I think we’re gonna see them overused very quickly. They’re very showy, very sort of character fonts; Lobster and Reenie Beanie are these two fonts — font names: don’t get me started. But Lobster especially I think we will be seeing on a lot of websites as sort of the logo text at the top and it’s going to become overused.
But, there are at least two-thirds to three-quarters of these are quite, I would call them plain and readable and workhorse fonts that you could very easily use for paragraph text. And there’s even a nice mono-spaced font, so finally at last we can get rid of Courier New for showing off code when we want to do a tutorial of some kind. I’m really looking forward to replacing all of the code listings across SitePoint with Inconsolata, which is the nice new monospaced font that they included in this collection.
Patrick: As you read these fonts it makes me think of dinner. What’s the house special? Well, Mr. Yank, the Lobster Reenie Beanie is quite delicious.
Patrick: And would you like that with a side of Inconsolata?
Kevin: (laughs) Can I interest you in a Molengo? Yeah, I see what you’re saying.
Does anyone else see any rendering glitches when they scroll up and down this page? I’m in Safari and Safari, while it does support downloadable fonts; they’re still kind of pushing the bounds, and I think having 20 of them in the one page is kind of pushing the limits because the dot on the I in Nobile has sort of fallen off the I and is sitting on the baseline between the B and the I. And when you scroll up and down a little erratically you tend to find text is only half painted in places.
Stephan: That’s an Apple problem (laughs).
Kevin: Yeah, it is an Apple problem and I think the fact that Google has released these 20 fonts is finally going to push browsers to fully — implement full and robust support for these downloadable fonts. It’s kind of been at a technology demo stage for the past year. Kind of the browsers have done just enough to make this feature available and they’re kind of waiting and seeing if it gets used before they invest any more time in it I would say. And Google is really hoping to tip them over the edge and force them to fully support this stuff.
Patrick: I’m in Firefox and I was gonna say the fonts actually look pretty good to me across the board. But then again on this main page they’re all at probably the 40/44 pixel size, so it may a little easier to look good at that stage. The Canterell font at the top though just looks a little sketchy to me, I don’t know, maybe that’s the style it is, but it’s like rough around the edges so I don’t know.
Kevin: Hmm, it’s pretty sharp. Yeah, you’re on Firefox on Windows as well, which for a while there had a reputation for horrible custom font rendering. So it’s really encouraging to hear they even look good in Firefox for Windows because that’s kind of a worst case scenario I’ve heard.
Patrick: Well, thank you.
Kevin: (laughs) Patrick! The worst case scenario.
Patrick: My computer setup is the worst case scenario, awesome!
Kevin: Maybe if you were running Firefox 2 still you would see worse results.
Patrick: At WordCamp Raleigh we were talking about how I still have yet to download Google Chrome, so.
Kevin: Oh, well. Speaking of Google Chrome, that was the third big announcement at Google IO, the coming soon in Google Chrome is Chrome Web Apps and specifically an app store for the Web.
Stephan, are you impressed by this as a fellow iPhone user who has been once or twice around the app store?
Stephan: Yeah, I think it’ll be a cool deal. I’m kind of looking forward to it a little bit.
Stephan: Nothing really excites me much more about the iPhone, you know, nothing comes out.
Kevin: It’s done.
Stephan: Yeah, this is kind of — this is kind of getting me a little fired up.
Patrick: Hot and bothered.
Kevin: The screenshots that TechCrunch have look kind of neat. I have to admit the idea that web pages aren’t things that you have to remember the addresses of and bookmark and visit, they become sort of icons that are permanently housed in the tab bar of your browser of choice, and you can launch these Web applications be they Google Mail or Calendar or Yahoo! Mail or whatever it might be; you just click these icons that float permanently in the tab bar of your browser, or at least that’s how Google is planning to do it in Chrome, and launch these as if they were applications installed in your browser. And whatever additional privileges these applications would need to do their jobs really well, whether they need access to your local hard drive to store cache files so they run quicker or work offline, whether they need access to other parts of your webpage or access to communicate with other websites, all that stuff they ask you for that permission when you “install” them in your browser rather than bothering you every time that one of those privileges like, for example, getting your Geolocation information, rather than bothering you every time. So it kind of looks slick. I’m not sure I want to turn my browser into an operating system, that’s the subtext here for me.
Stephan: And that’s kind of where I was thinking that this was going. Like is this the beginning of Google’s OS, you know, the Web OS that they’ve been —
Kevin: Chrome OS.
Stephan: The Cloud OS, Chrome OS, yeah.
Patrick: Google Chrome Cloud OS, version 1.2 beta.
Kevin: (laughs) Well, this is what they’ve said, they’re gonna release — as successful as Android is and as much as they were talking all about Android at Google IO last week, they have this other operating system they’re working on called Chrome OS, and they want to release these sort of netbook kind of devices that all they do is run the Chrome browser. So you open them up and instantly the Chrome browser pops up, and that Chrome browser is your OS. And so if you want to be able to install applications in any meaningful way on this new Chrome OS device then you need to support applications in the browser. So this seems to me to be those pipes that they’re laying in order to be able to have this Chrome OS be something of a full computer experience.
But how much is that going to affect the way the Web in general works? Is this just going to be a gimmick that you only use on Chrome OS devices or are we going to see new web applications being released as apps that you install in your browser not as sites that you visit?
Stephan: That’s blurring the line again, you know, of what’s in the web app and what’s a tool. I don’t know, I think we’re going to see people get in on this game for a little while and see how it works. I mean did they say if they’re gonna be able to sell them or anything?
Kevin: Yes, yes, developers will be able to sell them; the Google Chrome app store will just like Apple’s app store for their devices enable developers to charge for access to web applications. And Google will handle all the ecommerce and do a nice profit sharing split. And so, yeah, it is an easy platform for monetizing web applications certainly.
Stephan: Did you see the market? Because I know the Android market is one of the easiest — the Android market for developers is one of the easiest to use. I mean if it’s easy to sell a web app on this thing maybe people will pick it up.
Kevin: I think it’s early days, typical for Google IO it’s kind of a just barely got a tech demo working.
Kevin: So I think we’re a ways off from seeing the final design. But what they did announce is that, yeah, they’re planning to have a web app store where people can charge for their web apps or release them for free if they want.
I kind of like it as a concept. For a long time I would complain that we are — the fact that you can’t tell the difference between a web page that operates as a document versus a web application that operates as kind of virtual desktop experience just inside the sandbox of your browser. You can’t really tell the difference between those two things when they are just URL’s on the page or just links you’re considering whether to click or not. And that those types of content really should be separate in some way even if they are both considered part of the Web, we have this web of applications and we have this web of pages and they are interlinked, but the line was a little too blurry for me.
So this kind of is a nice step towards separating out those applications and saying alright folks, the applications are over here, they’re things you install in your browser; the web pages are over there, they’re things you visit in your browser. And I kind of like that concept. It worries me that Google is implementing this as something branded Google as a commercial enterprise in support of their browser on their upcoming operating system. It feels uncharacteristically closed to me. I wonder how much of this technology is open; whether a competing app store could open up and sell applications into the Google Chrome OS as well, and whether Google’s app store would be willing to sell into non-Google browsers.
Patrick: Can a competing app store open up and sell apps to the iPhone?
Kevin: Yeah, that would be nice.
Patrick: You know, it’s a weird thing because essentially what we’re talking about, and it’s a few different things, but a browser add-on, I mean we can call it apps, right, or you can call it add-ons or you can call it plug-ins or you can call it software; how many successful paid browser add-ons are there out there right now?
Kevin: Hmm. Yeah.
Patrick: I don’t know. You create a market, I mean the iPhone created let’s say the app store market, but people pay for the iPhone, people pay for their computer, but will they make the leap to say okay a browser add-on adds so much value that I will pay .99 cents or $1.99 or $2.99 or more? Or is it viewed as — will it always be tied to the Web like you pay to subscribe to a website, a gaming website, play games online or you pay to subscribe to this publication; is that the model? Is it essentially always a webpage tab? I mean you’re limited within the browser to what you can do. A browser will always be a browser unless it’s something different entirely, and that doesn’t appear to be the case here.
Kevin: Yeah. A good thought exercise I think is, for example, Google apps. SitePoint uses Google apps just like many companies do to host their email and shared calendars and all that sort of stuff. And we pay an annual fee to use that web application. And so there is an example of a successful web application that is charging for its service. But if you bring that into this Google apps marketplace, and you make it something that you go to in your Chrome browser and you install in your Chrome browser as a tab and the Chrome app marketplace says that will be twenty bucks for a year of access please, and you fork it over, then you go to your friend’s house or you open up your iPhone and you want to access that application not as a Chrome installed app but as a web application, are you able to? Because you bought access to that application through the app store, can you only access it through the app store? I’m hoping not, but there is this whole sort of tangly mess of user experience and expectation that they’re going to have to make really clear here for this to be a success.
Patrick: Yeah, I think and — I don’t know, I think it’s the Web, I mean I don’t know, I just see it a little simpler, I don’t know, that they’re going to have to have a web or, look it at like this, you have apps and you have the website; if there’s no website to login to if it’s not login based or some sort of validation then no you can’t access it through the web, it’s a piece of software like you buy a Windows application, you install it on your Windows desktop, and that’s where you use it. That’s the same kind of thing that would happen here, and really how many examples are there, let’s say, of an “app store” selling apps for varied platforms. I don’t think there are any, many, I don’t know.
Kevin: Yeah, even Adobe, before they did Adobe AIR as a free platform, if you want to charge for an Adobe AIR app good luck to you, that’s your problem to solve. But before that they had, oh, can anyone even remember what it was called? It was sort of the same technology that it was this bringing Flash and web technology to the desktop for applications, but it was tied together with an app store of sorts, that Adobe was going to let people sell apps in.
They gave me a tech demo of it sort of as an early press preview, and it was — I had the same sort of reaction that it sounds like you’re having to this Patrick, that it was like, oh okay, but is this really gonna work? People aren’t used to paying for that kind of thing. That was their big challenge is they were trying to make it cross platform, it was gonna be this Windows, Mac, wherever Flash goes this app store goes. It was I think Flex came after it, they kind of when this didn’t work they packed up and then they built their desktop widgets on the Web for Flex and now Adobe AIR is their second attempt to bring it to the desktop. And the piece that is clearly missing is the ability to sell Air apps commercially through some centralized app store. It seems like Adobe gave up on doing that and Google is gonna have a shot now, but will it succeed? I don’t know. Patrick I think your skepticism is well founded.
Patrick: As the least technically savvy person here, yes! Yes! My skepticism is founded, how about that.
Kevin: (laughs) One last thing before we wrap it up this week is the Twitter API is having massive changes in the next five weeks or so. It’s been awhile since we have fixated on Twitter. It used to be it seemed like you couldn’t even go through a podcast without spending half an hour talking about Twitter. And we’ve kind of — I don’t know what’s happened Brad?
Patrick: Now it’s HTML5.
Kevin: (laughs) Yeah, now it’s HTML5; HTML5 is the new Twitter.
Patrick: (laughs) That’s wrong in every single sense except for our podcast.
Kevin: (laughs) Brad, what is it about Twitter that we’ve kind of — it’s not hot anymore? At least not as far as podcasts are concerned.
Brad: Well, I mean it’s just not the new kid on the block anymore, you know, Twitter first came out in ’06 and it really started to get hot on the tech scene in ’07, and it was just every other day there was something new to talk about; their new apps, new API’s, whatever it was. And now it’s like, well, we all still use Twitter on a daily basis I’m sure, but it’s you know, it’s Twitter, it’s there and we use it and that’s what it is, so.
Patrick: Yeah, HTML5 is so hot right now. My mom just asked me about it the other day, I mean it’s that hot. Paris Hilton just threw it out there on Twitter; HTML5, that’s hot. Anyway.
Kevin: But, Brad, the irony is that I think there is more going on in terms of development of Twitter as a service now than there has ever been before. But people seem less interested. It’s like people are a lot more interested in theorizing and gossiping about something that isn’t progressing that we all wish it was, then once it starts moving it’s kind of like, oh, well that’s pretty boring.
Brad: Kind of like their ads.
Kevin: (laughs) What ads? Twitter’s running ads now?
Brad: Oh, the ads we’ve talked about. I don’t know if they’re up for everybody, I haven’t seen them, but the sponsored tweets is —
Patrick: Actually Twitter just said they’re gonna block — yeah, they’re gonna block any third party ad networks in Twitter. There’s been a few that have popped up, Adly, sponsored tweets, just I think it was the last couple of days they announced that those sorts of networks will be blocked from the platform because they’re gonna change their terms to make it a violation.
Brad: We should have another Twitter episode.
Stephan: Everyone’s just focused on Facebook and privacy issues and all that.
Kevin: Maybe next week guys; maybe next week.
Kevin: Patrick just sent around a really interesting graph from Compete; it’s the Compete graph from Twitter.com for the past year. And it was like crazy insane growth in the first half of 2009, and then it’s just leveled out and if anything in the past — in the start of 2010 it’s slipping down; the number of unique visitors that are going to Twitter.com.
Patrick: I mean it’s only the homepage, right, so what a lot of people would say well, okay, but what about the mobile use, how are people using it with their apps and on the phones and whatnot. But still, I think it would be reflected in that, so it’s telling if not completely precise.
Kevin: So the big thing that’s happening with Twitter is their changing their API, Brad?
Brad: Yeah, exactly. So, basically when the API first launched you could do basic authentication. It’s actually you literally just send username and password through like an HTTP request.
Kevin: So when I install like a Twitter app on my phone and it asks me to type in my Twitter username and password and I get a little nervous because I’m telling someone else my Twitter username and password, is that because it’s using basic authentication?
Brad: Well, it could be, I mean you never know exactly what that app is doing behind the scenes because you can also use OAuth which is basically what Twitter is going to start requiring in five weeks.
Kevin: So the idea of OAuth is once everyone is using OAuth you should never type your Twitter credentials into any page except a Twitter webpage. Because an app that wants access to your Twitter account will actually just send you to a Twitter webpage that says this app is requesting access to your Twitter account, do you want to allow it or not. And so you never actually have to give your Twitter password to anyone except Twitter. That seems to be the idea.
Brad: Yeah, I mean all of the old — you know when all the — Twitter was first getting hot, like all the little startups that integrated with Twitter, they would grab your username and password and it would essentially have to save it because every time you needed to access the API they would need your username and password to do so.
Brad: But yeah, you’re right, now they don’t need to do that. So you basically have five weeks if you have a Twitter-based service to update that from basic authentication to OAuth.
Kevin: I know when we first did something with Twitter on SitePoint it was one of our big sales; once you participated in the sale you could tweet about it to your friends, and we had sort of a Twitter graph, not a graph, but sort of a timeline down the side of people talking about the sale. And the first time we did it, it was all done through basic authentication; we just used the SitePoint Twitter account and we accessed the API using the SitePoint Twitter account’s username and password. And we had to jump through all sorts of hoops to get that account white-listed so that we could hit the API a little harder than a typical user would, and it was a real pain. And a year later when we came back to do the same sort of sale we needed to get that account white-listed again, they said no, we’re not gonna white-list your account, we don’t do that anymore, you should be using OAuth instead; that solves the problem. And it added an extra week for us to figure out OAuth because it is considerably more difficult than just sending a username and password with all of your requests. But once we did it, it solved all of those problems.
Brad: Yeah, there’s no — I don’t believe there’s any rate limits, right, once you’re using OAuth?
Kevin: The thing was that every user who visited the site would be giving us their Twitter — would be authorizing us to access their twitter account anyway because one of the requirements for participating in this particular sale was that you had to follow SitePoint, at least temporarily, in order to get access to the sale. So we would request through OAuth access to their account to verify that they were following us, and because we then had OAuth access to their account, then we they wanted to see our Twitter timeline they could do it through their account. And so we were sharing everyone’s rate limit instead of pounding on one particular account. It makes so much more sense.
Brad: Yeah, I know, like TweetDeck, for example, which is a popular desktop client, once they released a new version I want to say a month or two ago that introduced OAuth, and since then there are no more rate limits. So rather than only be able to hit the Twitter API 150 times an hour you can hit it essentially as many times as you wanted which was pretty nice.
Kevin: Hmm, yeah. And you know Twitter, the website, about this countdowntooauth.com has a sort of a countdown timer explaining exactly when they’re going to switch off the old basic authentication method, and they’ve got sort of list of reasons why OAuth authentication is a good thing, why it’s better. And one of the ones is that it improves their ability to scale and plan for heavy loads. So when we do a sale like that and our application is hitting their servers hard, they are able to see that application is the source of that traffic rather than just one particular account is misbehaving. So they can tell the difference between a popular application that they need to allow more capacity for and an account that has been taken over by spammers and is being abused. It’s just so much more robust a system and I love OAuth dearly as a result of that. But if I were a user who is using a Twitter client, a desktop Twitter client, that relied on basic authentication I might be a little angry; especially if the developer of that app was not actively maintaining it anymore because in five weeks’ time when they throw this switch, a whole bunch of old Twitter clients are just gonna stop working.
Patrick: And before we move off Twitter I just wanted to read from the post on the Twitter blog that mentioned the sponsored tweets adjustment.
Kevin: Oh yeah.
Patrick: They said — this was their quote, “We will not allow any third party to inject paid tweets into a timeline on any service that leverages the Twitter API. We are updating our terms of service to articulate clearly what we mean by this statement.”
So they don’t really come out and name anybody obviously, but it definitely sounds like, at least people are speculating, that it’s going to affect some of the paid Twitter advertising networks that are out there.
Kevin: Hmm. That is a vague statement; paid tweets, I mean, you know, I’m being paid to sit at my desk at work and tweet about SitePoint things, occasionally. Is that a paid tweet?
Patrick: What! What? I follow that account; this is an outrage!
Kevin: (laughs) Shocking.
Patrick: I don’t know. I think, well, they’re careful to say third party, third party injection. So, I mean I look at this and my guess, non-lawyerly guess, is that if I want to sell a tweet to somebody myself and put that in my stream that’s fine. But if I sign up with a network and then they inject it into my account because I’ve authorized them to do so through OAuth, etcetera, then that is where they’re putting — that’s what they’re putting a stop to. But that’s just my guess.
Stephan: We’re just one step away from them blocking Foursquare tweets, which would be awesome.
Brad: I thought I heard something about a filtered tweet feature coming soon to Twitter so you could do that.
Stephan: Thank you god (laughs).
Brad: That would be a nice feature.
Patrick: Filter HTML5, done.
Kevin: (laughs) In this week of Facebook privacy debacles I have to say I’m feeling pretty good about Twitter and trusting them to do the right thing, relatively speaking. Yeah, Twitter blocking the wrong tweets because they’re paid tweets is not really high on my list of worries for the Web at the moment.
So let’s finish off today with our host spotlights, guys.
Patrick what have you got for us?
Patrick: Well, if you haven’t heard, 99Designs won a Webby in the Best Web Service & Application of the Year category in the Webby People’s Voice Awards. And I hadn’t been following this really until they won; although I think I might have voted for them. But apparently they promised to do a special video if they did win, and that video is posted online on the 99Designs blog.
Kevin: Oh, they sure did promise. Uh, full disclosure: 99Designs and SitePoint are owned by the same parent company; just make sure that’s out there. But yeah, so all of the guys who work, guys and gals, who work at 99Designs work just like one floor beneath me, and we hang out a lot around the office. And they were going around going “We are totally gonna beat Dropbox and Tumbler, and when we do we’re gonna walk down the main street of Melbourne singing ‘We Are The Champions’.” And I think at the time I went yeah, right, you’re only saying that because you know you don’t have a shot of winning at all. And they won!
Patrick: And they did it. They posted a video lip-synching to ‘We Are The Champions’ by Queen. It’s about two minutes long.
Kevin: What! Lip-synching!
Patrick: Well, I take it that Mark Harbottle doesn’t have the voice of Freddy Mercury, but I might be wrong but, yeah, so they sing along to the track and walk down the street and just a good time in all, so check it out.
Kevin: Yeah. It’s a marvelous bunch of geeks celebrating and it is hilarious to watch, I highly recommend it. Just imagine them realizing they had won and then realizing they were going to have to do this, because that was a fun day around the office.
Patrick: And then get paid to do it.
Kevin: Yeah, well, of course.
Stephan what have you got?
Stephan: Well, since we haven’t talked about HTML5 enough today.
Kevin: It says that that one works on Firefox and Mobile Safari. And why is the Chrome logo kind of faded out there, I don’t know why.
Stephan: I just used it; I just tried it on Chrome, it works, so I’m assuming —
Kevin: Oh, it says nightly. Supposedly you have to be running a nightly build of Chrome for that to work. That’s weird.
Stephan: Yeah. So. It’s interesting though. And if you want to go out and play with like the Web SQL database storage stuff is pretty neat, so just go out there and play with it, it’s cool.
Stephan: And the guy’s name is Remy Sharp just if you’re wondering, so.
Brad: Mine’s actually it’s a couple weeks old, but it’s cool enough I definitely want to make sure we mentioned it. Face.com, which is a facial recognition technology startup, recently launched a public API and developer, an open API and developer community, and basically what this allows you to do is kind of tap into their facial recognition technology into your own apps. And it’s actually completely free; there is rate limiting, so it says you can do up to 200 photos per hour. So if you have a serious web app they have premium licensing, but essentially you can upload a photo, they have demos on their site, and it’s really neat, you can upload a photo with 50 different people in it and it will recognize all the faces; it also detects gender, it detects if the people are wearing glasses, it detects if the people are smiling. It’s pretty wild. It will detect heads if they’re on a tilt; you can actually hook this into your Facebook and Twitter API’s, you know, to work with the user’s social graph. So it’s pretty wild, but definitely check out the demos because you can upload or drop a URL to a picture and kind of see it in action, and it’s pretty amazing how accurate it is.
Kevin: I’m wondering if any web service would ever consider using this sort of thing as a login. Like, you know, smile at your laptop to log in. I guess it would have to be not a security critical application, but particularly on phones that would be really cool because I know one of the most painful things about surfing the Web on my phone is typing in passwords. And it would be, you know, if your phone has a front-facing camera for doing video calls, if you could just hold your camera up to your face and it goes oh yeah, that’s you, and logs you in, that would be really interesting. Maybe it would need like a sort of secondary — you also need a cookie to prove that you’ve logged into this site before or something like that, and it’s just sort of a re-login sort of feature. But that would be really exciting if this worked well enough. I know I have like face recognition in iPhoto on my computer, and I know that Picasa web service from Google also does face recognition to sort of group your photo library into people. And I don’t know what you guys have in terms of experience with that kind of thing, but my experience is it’s a little hit and miss, you know, you get a haircut and it’s suddenly very confused.
Patrick: I want to know how accurate the gender recognition is. Wish I could demo that right now.
Brad: It’s not perfect, in fact, —
Kevin: Patrick did it say you were a girl?
Patrick: I actually was trying to make that happen, but I can’t figure out a way — I can’t get the tools and demos to get going here, so.
Kevin: (laughs) That’s our challenge. Any listener try and fool it into thinking you are of the opposite sex. Tell us what it takes. You know, can you pose and make it switch your sex or does it require a funny hat? Let us know.
My spotlight for this week is a Video JS. We’ve talked about HTML5 video players before, and this is the latest one of the bunch. And the nice thing about it is that it is free and open source. Up until this point the nicest HTML5 video player that I knew, and the name doesn’t come to mind at the moment, but it was commercial so if you wanted to use it on your site you had to pay them a license fee. It was very nice, and these things provide slick sort of video playback controls that sit over the video. But the idea here is that they provide a video player that can play H.264 video in supported browsers without any Flash needed at all. And this is particularly important when targeting devices like the iPhone and the iPad which don’t have Flash video playback support.
<video> tag before the browsers can add it themselves. But, you know, they’ll maximize to the full browser window anyway, and the big differentiating feature that I find a lot of these free solutions miss is a fallback. So if you view this in Internet Explorer or Opera it’ll fall back to a flash player called FlowPlayer which is probably the most popular free Flash-based video player out there.
So there’s no down side to this as far as I can see. It’s free, it does everything you would hope for; the only pain here is that you need to encode your video in multiple formats. So right now it says it supports H.264, Ogg Theora, and the new WebM video format from Google as well. So if you encode your video in all three of those formats and point it at all three then you’ll get the maximum coverage. Presumably if you leave one or two of those out, maybe you only encode to H.264, I’m betting it will then fall back to FlowPlayer in more browsers, but at least you’ll still get full browser support. So a really nice library there if you need to host some video on your own site; give it a try.
And that brings us to the end of the show. Let’s sign off guys. Who are our hosts today?
Brad: I’m Brad Williams from Webdev Studios, and you can find me on Twitter @williamsba.
Patrick: I am Patrick O’Keefe of the iFroggy Network. Find me on Twitter @ifroggy.
Stephan: I’m Stephan Seagraves, you can find me on Twitter @sseagraves, got a mouthful there, and badice.com is the blog.
Kevin: And I am @sentience on Twitter. You can follow SitePoint on Twitter at @sitepointdotcom.
Visit the SitePoint podcast at sitepoint.com/podcast to listen to all our old shows and subscribe to receive new shows automatically.
You can leave comments on this show and we will endeavor to answer them on future episodes. The SitePoint podcast is produced by Carl Longnecker, and I’m Kevin Yank.
Thanks for listening. Bye bye.
Theme music by Mike Mella.
Thanks for listening! Feel free to let us know how we’re doing, or to continue the discussion, using the comments field below.