Louis: Alright, so here we are another episode this week. It’s my first time on the show with you guys so hi Stephan.
Louis: And hi Patrick.
Patrick: Hey, Louis, welcome to the show, the group show officially.
Louis: Yeah, yeah, well I’ve been doing a couple interviews but it’s my first time actually on the panel so that’s exciting. Unfortunately Brad could not make it today. We got pushed back a day later than usual because it was a public holiday here in Australia and I was not in the city so I didn’t have access to any kind of high speed Internet.
Patrick: We had some discussion amongst us if that was a real holiday or not because there was no holiday here, I’m not real sure.
Louis: Actually it was a weird one because there are usually two public holidays in April, so there’s Easter which is a normal one, and then there’s Anzac day which is kind of like our Memorial Day or Veteran’s Day here.
Patrick: So that was a terrible joke and I should feel really bad, I’m sorry.
Louis: (Laughs) no, but they actually fell on the same weekend this year, so we had a five-day weekend; the usual four-day weekend for Easter plus an extra one for Anzac day, so that’s why we had a Tuesday off which is kind of weird. But yeah, so I was out of the city but I’m back now and on the equivalent of a Monday morning for me I’m here recording the podcast with you guys.
Stephan: Glad to have you here.
Louis: Yeah, it’s good to be here. I realize I’ve got some big shoes to fill, Kev was a fixture on the show for so long, but I’ll try and step in and see how we can roll this, and as Kev said on the last show he’ll probably be around for some panel shows in the future so that’ll be fun as well.
Patrick: So before we get started with the stories I wanted to mention that the SitePoint Podcast will be back at WordCamp Raleigh for the second straight year, we made our first live show, our live debut last May in Raleigh, North Carolina for WordCamp Raleigh, and we’re going to be back again same dates, May 21st and May 22nd is the conference, we’re going to be recording the show May 21st from 12:00 p.m. to 2:00 p.m. Eastern Standard Time, that’s UTC minus four, and right now we plan to stream that through SitePoint.com/podcast, and there’s a lot that has to be announced or decided, but we wanted to make sure you knew about it so you could keep an eye out for that or if you’re in the area definitely stop by and attend the conference, that’s wordcampraleigh.com for more information.
Louis: Awesome. I imagine is Brad going to be giving some talks there or a talk?
Patrick: Yeah, both Brad and myself will be speaking. Brad is delivering a presentation called Surviving the Zombie Apocalypse Using Custom Post Types and Taxonomies.
Patrick: So yeah, that’s Brad’s zombie title, and then I’m talking immediately after in the same room doing Comment Moderation 2.01 which is topical for us on the podcast with some of the comments we get.
Louis: (Laughs) yeah, so I only checked my email this morning and saw that so that was a lot of fun. Alright so you guys ready to kick in to the news?
Stephan: Let’s do it.
Louis: Alright I think Stephan is up first.
Stephan: Well, I think the big story last week and into this week has been the Amazon EC2 outage, and the risk of the cloud, and that’s the article I have, it’s from PC World and it goes into some — brings in some questions about the cloud, and then I also have a VentureBeat article that talks about exactly what happened and what the issues were. I think it’s really about their backup system is how I read it, that they didn’t have as good a backup, Amazon that is, did not have as good backups as everybody thought, and when parts of their storage network started going out there was no rollover to the backups and people lost their websites including Quora, FourSquare, Reddit, things like that. So did you guys notice any outages? I didn’t really notice much because I don’t use any of those websites, so did y’all notice anything?
Louis: I think we had a little bit of latency here at SitePoint because we do use AWS services, but it wasn’t massive; I think most of the site was still online and functioning throughout it. It was, if I’m not mistaken, the outage was sort of restricted to one region, so the AWS East and a few availability zones within that, so that was one of the interesting things about this outage is it’s the first time that they’ve had, if I’m not mistaken, the first time they’ve had an outage that affected availability zones, so even if you had that multi-AZ setup which would usually protect you against outages you would still be affected in this case.
Patrick: Yeah, I didn’t notice much myself, just the news stories (laughter), it’s kind of funny because I don’t know, I’m not using the right services because I don’t check in on FourSquare, I’ve kind of forgotten about Quora right now. I did use AWS for serving audio to my music blog, but I don’t even spend that much time in that feature so I didn’t notice anything about it, and it is kind of surprising to hear that Amazon maybe didn’t have the backup system they should have just because obviously Amazon if they’re not on the ball then I don’t know if any of us are.
Stephan: There was a really good article, and I’ll need to find it and post it in the show notes, but there was a really good article on SAN storage, storage in networks in general and some of the issues they face. I’ll need to find the article because it went into great detail on how when you do backups sometimes if the backups are mirrored the problem is in the backup then it gets mirrored throughout all the different areas, and so when you try to restore you’re just restoring the issue. So no one really knows if that’s what the issue was or if, you know, because it took them a long time just to get things back online, so I wish Amazon would kind of come out and say what exactly happened and hopefully they will in the next few days because it’s worrisome for people who have websites with them or are running their websites with EC2 or any of the web services.
Louis: Yeah. There’s a couple of things that have come out in the wake of this. A number of blogs have come out and sort of said that one main article that I see linked to a lot of places is an article on the O’Reilly blogs by a guy sort of talking about designing for failure and how it’s possible to design around this kind of event and that The Cloud sort of in theory should be forcing people to think about their application architecture in a way that it’s not dependent on any one virtual resource or availability zone to keep functioning. And a number of people have pointed out either on Twitter or in blogs that one of the interesting things about this is to look not at the websites that went down like FourSquare or Reddit, but to look at the websites that didn’t go down; there’s a lot of people using AWS that were unaffected and a lot of people have pointed at Netflix as one great example of that. There’s this blog post on the Netflix blog from the end of last year, so from December of 2010, talking about — it’s called Five Lessons We’ve Learned Using AWS, and they’re talking about the way they’ve handled their infrastructure. And I guess the most interesting thing in there for me is their system called the Chaos Monkey, are you guys familiar with the Chaos Monkey?
Louis: So Netflix have this system which they say is one of the first systems that their engineers built when they started using AWS, and what it does is it’s a service that will randomly kill instances and services in their architecture, so at random it will just shut down an EC2 instance or kill an S3 bucket, and they use that as sort of a testing system to make sure that their architecture can still function if stuff goes away at random.
Patrick: That reminds me of the old SimCity games where you could just release a tornado in the middle of a city and just pop it down there and then see what happens (laughter) then rebuild.
Louis: Yeah, I guess that’s kind of the idea is they’ve set it up so that, I mean they’ve figured it out from the beginning okay we’re not going to assume that this stuff is going to be there forever or everything is going to survive so let’s make sure that everything will keep working. So it’s a lot of interesting stuff both in this article on O’Reilly and that Netflix article which was published last year so it’s not in response to this outage, but it’s interesting because it sort of covers what their approach to cloud computing is and it seems to have worked for them because they didn’t suffer any downtime at all because of this outage.
Patrick: I noticed a funny Tweet that Louis re-Tweeted from @jckhewitt who said: “OH in the office: It turns out the cloud is actually just some place in Virginia.” (Laughter) And this is a bit ironic because it was sort of a running gag on the show with Kevin and me where we don’t trust the cloud, and now this great opportunity and Kevin’s gone, so, you know, our fears have been justified! No, I’m just kidding.
Stephan: That’s actually what I was going to ask was are you a little more hesitant now to use the cloud, Louis and Patrick; I know Patrick you’re probably going to say definitely yes.
Patrick: I think it’s always important to have your own backups and all your own things that you control and not be totally in the cloud, you always need to have something to fall back on, a single point of any kind of failure is not a good thing.
Louis: Yeah. One of the things I guess that’s interesting about this I guess is that because Amazon is so huge and because they fail so infrequently, like they haven’t had a major outage in a very long time and I guess it comes across as a big shock to everyone because everyone just sort of trusted it like it was just this thing like it was air, like it was oxygen for the Web. But I guess this might have people thinking a bit differently about the way that they build their applications or their websites around cloud services and sort of understand that the stuff can fail and you have to have I guess fail-overs in place for when and if that happens; you can’t just put your whole business on top of someone else’s service and just assume that that’s always going to be there.
Patrick: Yeah, and maybe to exemplify that is there’s this thread on the Amazon Web Service Discussion Forums that’s got a lot of play and it picked up all over the Web about this company that is a heart monitoring company that they monitors hundreds of cardiac patients at home, and this post says that they were unable to see their ECG signals since April 21st, this post has been out April 22nd, so in other words they’re using the EC2 or the AWS servers for heart monitoring. I was very surprised when I heard that.
Louis: Yeah. Using the cloud in that kind of thing doesn’t strike me as weird, but it does strike me as weird that if you’re doing something that’s sort of that vital where you actually need to be monitoring people’s heart rates that you would architect in a way that has a single point of failure.
Patrick: Amazon goes down, people die, that’s the headline.
Stephan: So do you think we should question websites then that use — if you’re box.net and you use The Cloud services should you make that public and what your technology is, because maybe you don’t want to put your trust in that web service necessarily.
Louis: Yeah, I guess so. I mean a lot of these sort of bigger sites that have come out and said either oh we did something wrong, so one of the good examples I think is EveryBlock which was affected by the outage, and I’ve got a quote, I think it was on an article on Wired, on the Wired Web Monkey site, Paul Smith of EveryBlock they’ve got a quote saying, “Frankly we screwed up.” “AWS explicitly advises that developers should design a site’s architecture so that it is resilient to occasional failures and outages such as what occurred yesterday and we did not follow that advice.” So in this case he’s being transparent, he’s saying, you know, so a lot of these big services have blogs run by their engineers where they talk about the details of their systems and how they design for failure and how they architect their system, and I think being transparent with that stuff especially if you’re doing stuff that’s original and you’re actually, you know, because The Cloud is really young, so I guess there are a lot of people out there who just sort of assume that it works exactly the same way as hosting worked in the past, right, so you just get a virtual server and put your stuff on it in exactly the same way as you would with a physical server. But there are companies out there that are doing I guess more interesting stuff with it, so Netflix when they come out and post these are the details of how we’ve prepared ourselves for this I think it’s important for them to come out and explain what their approaches and their strategies are, and that’s people making informed decisions as well; if they’re not saying anything then maybe that’s a clue that they’re not totally up to speed.
Patrick: And I think that the explanation of whether or not it’s needed, and also the type of explanation you provide really depends on the service that you are, so box.net for example is this file kind of hosting service, it’s about keeping your files accessible, whatever those files may be, and collaboration and sharing things in The Cloud, and that certainly is about access, that’s a key selling point is that you’ll be able to access your stuff. So for them it makes a lot of sense to say this is way your stuff will be accessible, this is why your stuff will be secure, because we do this, this, this and this, because if they didn’t say that then people would wonder well how do I know that I’ll be able to access that document when I’m in Aruba on vacation, on Friday night at 2:00 a.m. or something. So that’s part of their business, where if you look at like a Walmart for example that’s not a conversation Walmart needs to have except in the nature of security; if people want to know why is my data secure with Walmart, why is my credit card information secure with Walmart, then that will be the conversation that they might want to have explaining how and why their website is safe, but no one cares about Walmart always being up necessarily other than to get and buy their stuff as soon as possible, so the different businesses kind of dictate that explanation and what it should contain.
Louis: Yeah. And I guess if you’re doing heart rate monitoring then it’s more important to get out there and explain how you’ll handle this stuff or not handle it as the case may be.
Patrick: Cool, so I can take the next story here, I found an article at TechCrunch that I found interesting from kind of a moderation, community, technology standpoint, the CEO of myYearbook, which is a social network, I don’t use myYearbook, do either of you guys?
Louis: I’ve never heard of it before now.
Patrick: Okay. Well, honestly I don’t think I’ve heard of this either, but they’re serving up to 750,000 video chats per day, live video chats, and so they have run into the problem that Chatroulette ran into which is for lack of a better term seeing a lot of private parts on their live video chat (laughter). And when the service launched, when myYearbook launched their live video chat, they found that one in every ten Chatroulette video chats was obscene, so 10% of all video chats was obscene, that might even seem a little low looking back on it (laughter), but they didn’t want that on their service. So they have kind of put together this system of monitoring their video chats that’s both human and algorithmic where they capture and analyze thousands of images a second from hundreds of thousands of daily video streams, and they sample the streams at random, and they have an algorithm that is based on many factors but one of the things they found is that an image with a face in it is five times less likely to contain nudity (laughter). Which makes sense I guess, but they don’t rely on just that because 20% of images that have nudity also have a face, so they use a lot of different things, chat reputation, social graph, motion in the video and other factors, and they pair that with a human element to take care of both false negatives and images that are simply decided to be safe, so they still do sample those images those images that are considered safe so that a human can lay eyes on it and really see if it is safe or not. And how they review that content is kind of a two-tier review system where if two reviewers find your behavior to be inappropriate you’re banned from the site, and those people are on staff 24 hours a day, 7 days a week, 365 days a year, and they have the goal of reviewing streams with a delay of no more than five minutes, so I almost think it’s an interesting conversation not only for moderation but for developers who are thinking of doing anything like this, the live nature of the Web, how there is kind of a responsibility to monitor your content.
Louis: Yeah, that’s pretty — it’s a big, big effort on their part. I guess not a lot of sort of startup style companies could afford to put people in front of that much video to moderate it constantly, but it’s interesting that they do have that first algorithmic layer of protection which might be really useful, I mean assuming this sort of technology becomes more common; if you’re just starting out it might be a good thing to just throw in some automatic moderation to start with and then once you get to a point where you can afford to have a person look through the — to clear out any false positives or false negatives.
Patrick: And just to throw some images, I mean not images —
Louis: (Laughs) I’m not sure I want to click on this link now.
Patrick: Just to throw some data behind this just to see how successful it’s been, they sampled a Chatroulette video stream, 1500 video streams from Chatroulette recently, and they found that 1.9% of them had some form of nudity in it, so one out of 50 basically. On their service they find one in 1,000 that have it, so that’s where they have found success, they’ve been able to limit that drastically. And another interesting thing that he pointed out is that he doesn’t necessarily believe that identity, like using Facebook Connect, plays much of a role in discouraging abuse because people did this on I guess there was an iPhone app called iChatter which you could be identified through and people still did this sort of thing. So, what he believes is having any sort of login at all discourages people, what they do is they actually fingerprint the devices people use to access the site through a service called ThreatMetrix, and then when they ban someone they ban both the user and their device so that they’re unable to access the site at all. So I thought that was interesting especially with the proliferation of Facebook Connect around the Web versus having more of a site-based login and you could have a little more control over.
Louis: Yeah, that’s pretty interesting.
Patrick: And so as a community guy, I know no one’s really interested in this story as you can probably tell, but there was one other quote that I felt was really accurate, he said that, “Based on their findings we believe purely algorithmic approaches to moderation will never provide adequate safety.” And I found that interesting because there seems to be on the Web this push to automate so many different things, and especially in moderation, but really you’re always going to need that human discretion to be able to really achieve a really strong level of safety and of guideline enforcement.
Stephan: That ties in kind of with the login thing, right, because he says as long as people can login they can be blocked, but at the same time you can’t block people based on an algorithm. So it’s these two, you know, he’s kind of taken technology two ways, he’s saying we’re going to force you to do one thing, right, and we’re going to also have human interaction. So it seems like a great approach to get rid of the crappy crap out there (laughs).
Patrick: Right, because no one can be banned based on the algorithm, even the people that are flagged it takes two editors to say okay, yeah, that’s that thing we don’t need to see and then you’re banned.
Louis: Yeah. Well that makes sense, it’s interesting because they’re dealing with something that’s just video so it’s very black and white whether someone is showing something inappropriate in a video or they’re not, whereas when you’re dealing with comment moderation sometimes it can be hard to draw the line whether someone’s being inappropriate or being abusive, I mean it’s easy to algorithmically detect spam if it’s just gibberish with links in it, but when someone’s just kind of being a dick that can be hard; if we come up with an algorithm for detecting that I’d be very impressed.
Patrick: Right. Well there you go, that’s the next idea. I’m sure a few developers listen to this show so get to work on that.
Louis: Yeah, that’s what we need, that’s the next thing. Alright, well I’ve got a story to throw in here; it will be my first contribution to the panel so hopefully it goes well. So the story I came up with was a blog post on Smashing Magazine from just this past week, and the title is CSS3 versus CSS, a Speed Benchmark. So we hear a lot about CSS3 and a lot of times what we hear has to do with just how much fun it is for developers to be able to do this stuff that used to be harder to do, so things like rounded corners or drop shadows or text shadows or gradients or anything like that, that used to take a lot of tweaking around and playing in Photoshop that now you can do easily with CSS3, yeah, okay that’s great. But what the author of this article has done is decided to try and do an actual comparative speed test, sort of a somewhat scientific approach to this process. So he’s come up with this fairly straightforward design for a one-page site, and he’s coded it up first of all using CSS3 so using new techniques like border-radius and box-shadow, and then he’s gone through and coded up the entire site again for the exact same look but using more traditional techniques, so using background images to fake rounded corners or fake drop shadows. And what he’s done is he’s come up with comparisons both in the time it took to code and on the load time of the two sites, and the results are pretty impressive. So first of all when he codes the site in CSS3 it takes 49 minutes to design up his site using CSS3, and using CSS, the regular image based method just takes 73 minutes, so it’s about 30% faster. And then to do with the file size there’s also a difference where CSS3 total file size download is about 10% smaller and about 45% fewer HTTP requests required because you don’t have all those little separate background images for the corners and such. And the results in terms of the load time, the CSS version loads in 3.3 seconds and the plain CSS version loads in 4.7 seconds, so above and beyond what we’re used to thinking about in terms of the advantages of CSS3 being it’s more fun to work with there’s also some really, really practical advantages in terms of page speed.
Patrick: That’s pretty impressive across the board.
Stephan: The speed differences in development time are huge and speed differences for load time are somewhat smaller and you’d think that it really scales, I mean I wish this was done on a bigger scale I guess, because it seems like it’s very limited and it would be negligible to a person loading the page, you know what I mean?
Patrick: Will they notice these fewer requests and fewer size, I don’t know, I’d have to see the browser studies, but I wouldn’t think that, you know, it’s probably one of those split second type of deals where only people who are hyper-aware of it would notice; you think that’s fair?
Louis: There’s an impact; so the article links to another article on Steve Souders’ webpage where he’s talking about the impact of performance optimizations like this on sort of traffic because you just get fewer bounces and a few extra clicks, so he’s got some statistics from search engines who did some studies on this. So if you look for example Yahoo did a study where they found a 400 millisecond slowdown resulted in a 5 to 9% drop in full page traffic. Shopzilla did a study, well, they optimized their site, they sped up their site by five seconds, increased their conversion rate by between 7 and 12%, Mozilla shaving 2.2 seconds off their landing pages increased downloading conversions by over 15%. So there is some data out there that would seem to indicate that these what seems like small differences, and it seems like if you’re trying to see okay which one loads faster and I load them both in my browser and they kind of look the same, like it’s a difference between what is this about 3 ½ seconds versus 4 ½ seconds for these two versions of the site? But the data seems to show that even those differences that can seem imperceptible to you will still affect the way you interact with the site even if it’s on a maybe not fully conscious level.
Patrick: Yeah, and I mean three to five seconds is actually a pretty sizable amount, and maybe the answer is that maybe this changes or these changes won’t necessarily push you three to five seconds, but if you’re starting to optimize your pages then they can definitely be a part of that speeding up process.
Louis: Yeah I guess it’s also fair to point out this is a fairly simple website, it’s a one-page thing, it’s got a few images on it, most of the weight comes from this one big — so it’s a sample website for Mercury automobiles, it’s got this very retro 50’s look to it. But most of the weight of the page comes from this one big honking car image in the center of the page, so you know tweaking a couple of rounded corners here and there or a button image that you don’t need to load in CSS3 because you’ve got gradients and rounded corners, it’s pretty minor, but if you had a bigger page that was using a lot of these different elements here and there you might see a bigger result.
Patrick: Another story that crossed my browsing ritual I guess you could say is the launch of Facebook Send, now you can’t actually add a Facebook Send button to your webpage, not yet, there is a limited pool of about 50 sites that are participating with this feature but essentially what it is, is it’s just like the light button except it says send, first and foremost, and also it kind of gives you the ability to send to a specific group of people, a specific person on Facebook or a specific email address. So it’s just like if you were sharing a link or sending a message to people on Facebook, you know how you can type someone’s name and it will fill in the rest, you can type a group you’re a member of on Facebook and it will fill in the rest, that’s essentially what this feature allows you to do. Think of it as email to friend but more powerful where you can send to people on Facebook and also send to email addresses and share a link with them and also use the Facebook platform. I don’t know how successful email to friend features really still are these days; does SitePoint still have an email to friend feature, Louis, on the articles?
Louis: Great question, I don’t even know the answer to that question. We did a fairly recent redesign of blogs so I don’t actually know, I can try and pull that up and see.
Patrick: Yeah, the sharing images as far as I can see are just Facebook and the like button, the Tweet button and then the ShareThis icon.
Louis: Yeah, I think it’s possible that the ShareThis has an option for email, so it might still be there. Yeah, but I guess email this is pretty common on sort of mainstream news sites, you see it a lot, and I think for those types of sites it’s probably pretty common because a lot of people are at work and they may have for example Facebook blocked, but if they want to send around a link they can just use an email link. But I think it will definitely see some uptake because everybody loves Facebook, right?
Patrick: Right. And the thing about it though that I noticed when I look at the two icons next to each other on this article is the Facebook like icon which just allows you to like a story on your Facebook wall, so that’s kind of the difference between the two, but it has the Facebook icon on it which may seem simple but that goes a long way toward getting people to actually click that button, because they’re on Facebook they know what Facebook is. The Send button just has like a little talk box, talk bubble I should say, in it where it looks like someone’s saying something like it’s a chat or a comment page almost, not necessarily send which might be more tied to email, an email icon, or a message, an envelope, so I don’t know if that’s going to hurt it just from a design perspective, but I guess it will be interesting as people play with it to learn how effective it is.
Louis: Yeah, that seems like something that they might wind up changing down the line because I think for someone who’s a Facebook user, and I’m not really a heavy Facebook user, I have a page but I don’t really use it, for someone who is a Facebook user it kind of makes sense; you see that Facebook icon and you’re like oh I can post this to my Facebook and I just hit like and now I don’t even have to think about it. But when it’s something like Send unless they do — unless Facebook does a sort of campaign to make their users more aware of this I don’t see how people would really tap into it and say, oh, that’s a way to send to people on Facebook.
Patrick: Yeah, so it’s almost like it’s going to take some time to catch on. I don’t know if I see this catching on as quickly as the like button.
Louis: Yeah. I can see the appeal from the point of view of web developers or people who run sites, especially if you previously have your own email to friend feature, and that takes time to develop, and if this can do email as well as sending Facebook messages then it sort of gives you that functionality for free even if people aren’t Facebook users. Does it work if you’re not signed in to Facebook, can you send it to — can you use the email functionality even if you don’t have a Facebook account, do you know that?
Patrick: That is a great question, and no I don’t think I do know that because I don’t know a whole lot of information available about it right now. The article at TechCrunch by Jason Kincaid says that you can share obviously to an email address but do you have to be signed in to Facebook, and I’m going to speculate here, total speculation on my part, that they will allow you to do that. I don’t know for sure but it’s easy enough where you type in an email address here and they can allow you to send it, now on the other hand, on the other hand I’m going to contradict myself, maybe not, because maybe being logged into Facebook is sort of even a spam prevention mechanism where they can tell, hey, you Stephan Segraves, you are logged in and you emailed this person, they reported that as spam so now you’re in trouble with us, that sort of thing, whereas if it’s just an unrestricted thing obviously that presents sort of a privacy, security type of issue.
Louis: Yeah. I think it’s interesting in both senses. Like if they do let you use it just to email then that could sort of speed up adoption by site owners because it gives you that email to friend feature for free and you don’t really have to develop it yourself or do anything other than pop this Facebook thing. But on the other hand it seems like that’s not in Facebook’s interest, so in their interest they want everyone to be logged into the Facebook ecosystem all the time, so they might sort of pop you a prompt if someone clicks on send a friend, oh you can send this to your friend you just have to login to Facebook first which seems like what the approach they might more likely to take.
Patrick: Yeah, and actually you know what I’m going to test that right now. Let me log out of Facebook, I’m logging out right now and I’m on an Orbitz page which is one of the pages they have, refreshing it, so let me see if I can; I’m going to guess I can but let’s see. I’m logged out of Facebook, I press send and it wants me to login so there’s your answer, no you can’t send it without being logged in.
Louis: Well I guess that’s interesting right, and we struggle with this here at SitePoint, you know whenever we do a new page or a new promotion or something we have to, you know, on the one hand you get a lot of uplift from those Facebook integration features, like when we did our December sale last year our whole comment system for the thing was powered by Facebook comments and it generated a lot of uplift, but on the other hand you want to be careful and not tie your whole thing too tightly in with these services, so it’s kind of a bit of a tightrope.
Patrick: Yeah you don’t want to give them too much power over you or you end up with Google which we don’t want. No, I’m just kidding.
Louis: Yep, definitely interesting but I don’t think it’ll stop our concerns about these, I think we’re definitely in a minority with regards to this stuff. I always find myself this handwringing about oh Facebook’s taking over the Web or Twitter’s taking over the Web or whatever and nobody cares, they’ll still do it.
Patrick: Yeah, then you realize it’s really to .0001% of people that actually really care about those types of things, the rest don’t know what a search engine is, so yeah.
Louis: (Laughs) yep, so they’ll just keep clicking those like buttons and maybe those send buttons. Although I do think that that’s — the absence of the Facebook logo is a really interesting one, but maybe who knows, maybe they’ll get to the point where they can assert trademark over the word send.
Patrick: Yeah or maybe they’ll just use a little envelope with a very small Facebook button on it, maybe that’s the solution.
Louis: So one more story this week that I found I think was just this morning, The jQuery team announced the release of the beta version of jQuery 1.6 so that’s of course interesting to just about everyone who does web development because just about everyone uses jQuery these days. Looking over the changelog really quickly it’s notable that this is probably the first release, or the first beta release of jQuery that hasn’t included any really show-stopping features, most of what’s on the changelog there is really just performance enhancements or bug fixes to do with specific browsers behaving in different ways and they’ve cleaned things out a little bit and rewritten a few methods for consistency, but overall I’m not seeing anything really major in that so obviously it’s good news for developers because it’s always good to have new versions of this stuff, but maybe jQuery’s just reaching the point where it’s so mature that there’s nothing really significant that needs to be added to it and it’s just all about fine-tuning from on out.
Patrick: And there’s a disappointed kind of expression in your tone there, but it’s not a bad thing is it that it’s so mature?
Louis: No, it’s not.
Patrick: There’s no new features in this thing!It’s turned into IE6.
Louis: (Laughs) man, that’s gonna bring on some fierce commenting. Yeah, jQuery’s the new IE6, let’s make that the title of the show and see how that goes over (laughter). But yeah I guess you always kind of want cool new toys, right, and if your toy has gotten to the point where it does everything you can think of wanting it to do and there’s nothing really to add other than to make it a little bit shinier and a little bit faster it’s a little bit less of a Christmas morning feel when you unwrap the new version of jQuery, right?
Patrick: Yeah it’s more like a day after Christmas.
Louis: Yeah, there’s leftovers in the fridge. Alright, well I think that’s a wrap for the news this week, so I guess we can move straight on to the host spotlights, Stephan do you want to start us off?
Stephan: Sure, so here in Houston we have a university called Rice, it’s a really large school and a very well known school in the state, and these guys are doing some cool stuff to extend Wi-Fi up to a mile using empty TV channels, and so there’s an interesting article on Ars Technica about exactly what they’re doing and how they’re doing it, and basically they’re tapping into TV antennae and using them to broadcast a Wi-Fi signal over great distances. So we have a free Internet service here in Houston called Technology for All, and some people live on the outskirts of it and these guys have figured out a way to extend the range of their signal stronger, so very cool stuff.
Louis: Hmm, that’s really interesting. It’s surprising that you’d think that at that point maybe, I don’t know anything about radio tech, bit of a disclaimer here, but you would think that at that point the limitation would be the antennae in your device, right, not the base station, but my phone or my computer still has to send data back to that base station so, you know, I guess I’m curious as to know how that works.
Stephan: Well they’re doing it both ways, so the TV station that they use then ties in at that TV broadcasting tower into the WiFi network directly so they’re actually hardlined into the Wi-Fi network at the TV towers, so that’s actually probably what’s making it a little bit faster too.
Louis: Yeah, definitely. We can maybe look forward to improved public Wi-Fi in the future.
Patrick: So my spotlight is New Media Accelerator, that’s newmeconference.com, New Media Accelerator is an accelerator for startups aimed at minority led startups, and they are going to house their conference this summer in San Francisco. One of the people behind it is my friend and friend of the show, been a guest a couple of times, Wayne Sutton, who’s always working hard to encourage diversity in the tech space where it’s much needed, and it’s just a really cool event out there for any startups that are led by a minority you should definitely check it out and go out there and present your startup and then work with their speakers and mentors, I think it’s a great opportunity.
Louis: Yeah, it’s great to see people doing kind of innovative things to mix things up in the tech world which is, it’s surprising to think of tech startups as being kind of sort of this stagnant area, but sometimes it does feel that way and sometimes it feels like you need a bit of fresh air and this kind of thing is great for that. Awesome. Alright, my spotlight is a post that was written by Alex Walker who is our in-house designer here at SitePoint, he wrote a post a couple weeks ago that many people might have seen, but in case you didn’t it’s called the Cicada Principle and Why it Matters to Web Design, and what it is is basically this technique that Alex came up with using multiple background images, so kind of circling back around to the CSS3 thing, so CSS3 let’s you set multiple backgrounds on a single element. And what he’s done is using semi-transparent PNG images with widths that are prime numbers, I realize this is getting kind of complicated, but that way you can create these repeating tile backgrounds that really don’t look like they’re repeating because you’ve got this base layer that’s say 17 pixels wide and a layer on top of it that’s 13 pixels wide, and because those never fall into sync you can create these seamless tile backgrounds that seem to go on forever where you don’t notice a pattern. Because we’ve all had this experience where you’re on a website and you’re looking at the background and it’s a tiled background and you can kind of see, you can see the pattern repeating, right, you can see where the tile is and where you’re seeing the exact same thing over again. And there’s a couple of cool examples up there of backgrounds that he’s made and they’re really impressive because they really do look like they’re seamless and they can be done with not a lot of kilobyte weight in images because they’re all individually very small images. So, yeah, that’s my spotlight and they’re currently running I think a little bit of a competition if people want to try and submit their own there’s a bit of a contest going and some prize for that, so worth checking out.
Louis: So that’s a wrap for the stories and the spotlights this week, do we want to go just quickly around the table.
Patrick: Absolutely, but first applause for your first episode (laughter). Oh, my God, wait a minute, redo, redo! I’m sorry. I said Lewis again.
Louis: I survived! (Laughter)
Patrick: But first I just wanted to say great job, Louis, on your first show, it’s great to have you on board, yeah, get through those nerves and we’ll work through it, right?
Louis: Yeah, appreciate it.
Patrick: Good job, Louis, it’s great to have you on the show. My name is Patrick O’Keefe of the iFroggy Network, you can find me on Twitter@ifroggy, i-f-r-o-g-g-y.
Stephan: I’m Stephan Segraves, you can find me on Twitter@ssegraves and my blog is badice.com.
Louis joined SitePoint in 2009 as a technical editor, and has since moved over into a web developer role at Flippa. He enjoys hip-hop, spicy food, and all things geeky.
7 Habits of Successful CTOs
"What makes a great CTO?" Engineering skills? Business savvy? An innate tendency to channel a mythical creature (ahem, unicorn)? All of the above? Discover the top traits of the most successful CTOs in this free guide.