I just got done replacing my iPhone screen, which I shattered over the break. The part was $30 on Amazon, as compared to a $130 third party repair, or probably a $200-300 Apple/AT&T repair. Yes, I am feeling pretty satisfied right now.

There are plenty of good videos and resources out there, so I won’t add my own, but would just like to say—yes, it can be done!

This past week, one of my students presented his first paper at HPDC. There is a more detailed blog entry at the research website by the student, but I wanted to mention it here too. (Post: http://www.macesystems.org/2011/06/incontext-simple-parallelism-for-distributed-applications/)

The very short story: the Mace toolkit has scalability issues since events must run atomically (think a big lock protecting events to run only one at a time). This paper describes the first step towards loosening that restriction, and running different events in parallel as long as they are not both trying to write to global state.

This post was promised some time ago, about our paper published at the conference on Foundations of Software Engineering (2010), a top conference in software engineering. Instead of posting it here, however, Karthik (one of my student co-authors), posted a description of our FSE paper here: http://www.macesystems.org/2011/04/finding-latent-performance-bugs-in-systems-implementations-fse-2010/

The very short description: by building robust systems, we hide some of our correctness bugs, converting them into performance problems. Our paper is about using model checking concepts to discover such bugs with a minimum of developer effort. See the post and paper on the group website.

So a couple weeks ago our gas grill stopped working. Right in the middle of cooking. Originally, I thought we were just out of gas, so I had swapped it out with our spare tank we keep on hand for just such an occasion. But I could not get the grill to re-light. I couldn’t hear the gas either, which I usually can do. I re-tried many times, before finally giving up. Later on, after doing some debugging, I had decided to try replacing the regulator. Unfortunately, despite using a wrench, etc., I could not disconnect the old regulator/hose assembly. At this point I was fed up. The grill was I think about 6 years old, so it lived an okay life.

In deciding what to do about it, I did not envy the thought of replacing it with another gas grill. Gas grills just seemed to me to be overly complex, with a large number of parts which can break and stop working. Couple this with the fact that I’ve never been as happy with the gas grill as I wanted to be — the burners worked pretty well on low and high, but anywhere in between, and they would flicker out and back on, which was always puzzling. Plus, I have to say I never felt like the gas grill made foods taste all that “grilled”. I’m not sure how much benefit it had over the gas range which I had indoors. That flame was consistent, reliable, didn’t run out, was self-lighting, instantly ready, and did not require me to cook in the outdoor elements. So, my go-to grill was the cast-iron grill tray on the gas range, or sometimes, the George Foreman.

So, in looking at new charcoal grills, they were cheaper, promised more-grill-tasting food, fewer parts to break. After all, in the end, it’s basically a fire-safe kettle you put charcoal in. The Weber grill I got had the added feature of a one-touch ash-sweeping system to make it easy to keep clean. It took all of 30 minutes to assemble, and the only tool it required was something to tap the caps on to keep the wheels from sliding off.

Pros and Cons are pretty well established across the internet, but here are a few considerations I went through:

• Cost: Gas grills are more expensive, but the fuel is supposedly much cheaper. If I the grill is $200 cheaper, and you grill 20 times a year, and your grill lasts 6 years (like my last one), you would have to save $1.66 per meal to make the difference up. I’m not saying it’s impossible, just that I’m not convinced the cost is all that significant either way.
• Convenience: Gas grills are instant-on, while charcoal grills take more effort and time to get started. Technically, this one is true. However, the starter on my gas grill didn’t last long, and then I needed to use a lighter to start it. Next, I never knew how much gas was left to know when it might run out. Also – as for the instant-on: I always needed to clean the grates before use, so it wasn’t like I could start cooking immediately (not to mention pre-heating the grill). Now, perhaps the charcoal takes longer, but since I generally use that time to prep the food, it seems to be about the same amount of pre-prep in both cases.
• Taste: The charcoal grill definitely is more grill-like in it’s taste. Plus, I can add fresh herb sprigs to add some smoke to the cooking. Very cool.
• Cooking simplicity: I have to say, the actual cooking seems pretty nice. Putting the lid down, and just waiting an amount of time for things to be done is great.

I’ve used the grill 4 times now. The first time was an utter failure (I didn’t figure out how to start it properly). Then we had friends come over to show us how it works. Since then I’ve grilled twice, and both times came out great. So, I’m now a happy charcoal grill operator now.

Does anyone want two propane tanks for the exchange system? One is full. The other has an unknown amount of propane (could be empty). I won’t be using them anymore.

Generalizing this out, it is basically illustrates the principle of keeping it simple. The gas grills are more complex, more features, more money, and should be easier and better to use. But for my use, the charcoal grills are more durable, reliable, predictable, and therefore useful. Software often follows the same principle. When we add a lot of complexity to it, we generally add bugs, issues, and make it much harder to know what the software will do. So in summary: keep it simple. Even if you lose some features or flexibility, you may make up for it in the bigger picture.

I have been a long-time fan of Google, and I generally find their products convenient, high-quality, useful, and innovative. Unfortunately, with all that useful product has come an ever-decreasing amount of privacy. It is now possible to do much of your daily computing work with entirely Google products — email, web search, documents, maps, chat, voice & video chat, youtube videos, picasa images, and now even DNS. Beyond all of this content which users freely give to Google, their analytics allow them to track you even further as you venture into non-Google sites, since these sites often use Google to provide ads to their sites, which allows Google to see what other sites you visit.

Recently, Eric Schmidt, CEO of Google, responded to questions in a televised interview which suggested a lack of concern for the preservation of privacy (some posts about it link to the video of the interview mozilla ceo’s blog urging switch to bing notes from the electronic fontier foundation summary of various posters). I am a little torn about how to react to this. On the one hand, I admit that the statement is factual — Schmidt is telling people that Google and other companies store this information, it is all information made available to Google, and Google is legally obligated in some cases to make this information available to authorities. On the other hand, I think it shows a complete disregard for the importance of user privacy, and the need for companies who are given such information to be good stewards of that information. Simply hiding behind the Patriot Act as a way to defend your practice of gathering, correlating, and then divulging information as desired may be legally sound, but does not reflect the attitude I wish the company gathering information to be taking.

Consider this: if you use Google DNS, yes, you may get some performance benefit over the DNS provider from your ISP. But, by telling your computer to use Google’s DNS servers, you will effectively be sharing the names of all websites you visit with Google. And don’t think for a minute that their automated mechanisms won’t correlate this with both the searches you are using to Google and the data of all Google accounts you maintain. (Note that a naive objection here is that you otherwise give this to your ISP. But your ISP already is transferring all of your network communications, so that information will be available to your ISP whether or not you send them your DNS queries).

From Schmidt’s point of view – you should already be aware of what information you are giving to Google, and the fact that Google may pass it along to others. But are you? Google could pick things out of emails you send or receive, match that up with the GPS data from your cell phone that you pass to Google so it can place you on the map, the searches you conduct on your phone, and the hostnames of sites that you visit, and build a very precise picture of what you’re doing. Granted, certain things in that profile might have been misinterpreted, and that profile could make it look like you’re doing something very embarrassing.

You see, it’s not the fact that Google has or is collecting this information per se that’s bothering me, but rather that when making this statement, Schmidt doesn’t say “Yes, all of this is true, BUT here are the steps Google is taking to safeguard users…” No, Google’s business is enhanced precisely because they can build these profiles of users.

Google’s long standing reputation of following their motto of don’t be evil is eroding. Thus far, I wouldn’t necessarily say they are being evil. But I no longer feel I can endorse them whole-heartedly as I used to. You should think seriously about what tools and products of Google’s you are using, and what kind of information you are giving them, and whether or not you really want a company who does not proclaim their desire to defend your privacy to have all that information about you. Does the cost outweigh the benefit of using their products?

Subtitle: protecting the language

So I am often bad with sayings and grammar; as a result, I am fascinated by the subject. There are a number of these which I have often misused, some of which include:

• I’ve used “here here” when I should have used “hear hear”. See this post for details.
• I’ve said “intensive purposes” when I should have said “intents and purposes”. I had even worked out a definition, including how they differed from “extensive purposes”.
• I’ve said “mute point” when it should be “moot point”. See here for details.
• I also mix up a variety of sayings, such as the nonsensical “double-bladed edge”.
• And the list goes on…

In addition to the language rules blog linked above, another great place to learn about such things is A Way With Words, broadcast on many local NPR stations.

Unfortunately, I see these mistakes as muddying up our language, so I would like to get better about it. But rather than
going out and searching down each of these cases to learn independently, I feel like there is a better solution. As a computer scientist, I recognize that we already have the perfect mechanism for this — the grammar checkers. Our word processing software already has a grammar checkers, and our web browsers have spell checking, so it may just be a matter of time. I think the grammar checker should be adapted to look for misuses of the language, and to suggest alternatives to writers. This suggestion should come complete with internet links to learn more about the cases found. So in the future, when I “tow the line”, it can let me know that I should instead “toe the line”. So short of listing all commonly mistaken sayings, how can we build software to do this? That is my question. But in the meanwhile, I would just settle for a [possibly community maintained] database of common mistakes it can check for.

So at the Journal and Courier online (thanks for the recent feature as blog of the week!), they run a daily story about “pending obituaries”. For examples, just do a quick search.

I, for one, find this practice very amusing. It is particularly amusing when it shows up in the optimized page for hand-held devices as “breaking news.” I mean, judging on the title alone, it seems like this should refer to people who are about to die, but who haven’t yet. Something like “John Doe, who was hit by a truck this morning, is in critical condition at the hospital. He is expected to pass within the hour.” Instead, it seems to be brief announcements of the completion of life, but without the full details. (These articles inevitably contain a sentence at the end referring the reader to the next day’s edition of the J&C.)

In the future, may I suggest (to no-one in particular, since I’m not making a point of sending this to the paper) that we find a more appropriate name for these articles? Though I have to say obituaries in progress or partial obituaries aren’t any better. Maybe we should change the word obituary, since it tends to suggest that it should have full details. Perhaps we could call them death notices instead? Then refer readers to the full obituary in the print edition. Or, maybe just leave it as it is. Maybe we need a sense of humor about the passing of our lives.

Tonight we were at the LeAnn Rimes concert, in which there was a posted no-cameras sign at the entrance gate. As you will see later when I post 2-3 of my own, you might expect that this rule is not well enforced.

So what’s the problem? Of course, the problem is that you can’t buy a cell-phone anymore without a camera, just about. Thus, unless you are going to either check every cell phone, or not allow them at all, you aren’t going to keep cameras out of the concert. (Of course, these cameras are also small, so unless you plan to use a metal detector, you probably won’t notice the cameras [cell phones] anyway.)

Thus, it follows that there were a LOT of people taking pictures at the concert with their cell phones. Throughout the concert they would walk in front of the first row (between it and the stage) and pause long enough to snap a picture. This of course was very annoying for those of us in the first few rows. There was also no attempt on the part of security to prevent or curb this activity. After all, what are you going to do, short of making people cross the venue at the back?

So accept as a given that people will have cell-phone cameras. It no longer makes any sense to prevent the use of the vast majority of consumer-grade cameras, which are only marginally better than the current generation of cell-phone cameras. And not surprisingly, there were people using those as well. Oh, and the bigger ones too—no, not so big that they were bigger than someone’s head, but still quite big. People were also not shy about it as you might expect, quickly snapping the photo and then hiding the camera so as to pretend they didn’t take a photo. No, they would walk right to the front with their quite-obvious-camera, and take a picture, complete with flash. Oh heck, why stop at just one. Get another one while we’re up here, in case the first doesn’t turn out.

The woman seated just in front of and to the side got a bunch of pictures — many quite good (I know, because it was impossible to avoid watching her LCD screen as she setup the shot, took it, and then checked its quality. It was, after all, being held over the level of all our heads while she did so to avoid anyone in the audience being part of the photo).

So, I think the time may have come to abandon the no-cameras rule, since it is so clearly not actually applied. Instead, we should be thinking of ways to make the cameras less obtrusive during the show. Perhaps have a place people can go to shoot their shots which is out of the way of the main audience. Perhaps have a song break where you tell everyone in the audience to get their photos out of the way now, and then ask them to put the cameras away and enjoy the concert. Coming from the artist themselves, seems more likely to be heeded anyway. Perhaps tell the audience they can take photos but to NOT use a flash. Some of them still will, but if you couple it with a reasonable explanation of why you shouldn’t use a flash, I think many people will respect it. Also, have a big sign posted indicating that shooting photos is acceptable for personal use, which still lets you crack down on those trying to make a buck off their concert shots, while allowing those people who you really can’t stop anyway shooting a photo to post to Facebook.

After all, those Facebook photos are probably doing more to promote and benefit than they are to harm.

[Review: it was a good show, and the third-row center seats were excellent.]

My wife and I have recently been discussing the idea of canceling our cable TV. There are a variety of reasons for doing so, which include these circumstances:

• Cable TV is trying to push digital cable by removing channels from analog cable.
• We see no present value in the additional costs of digital cable. In fact, we see no value in getting bigger, sharper, TVs, as we feel the picture is just fine, and sufficiently large to see from our sofas a mere 8 feet away. So it’s not about quality. And it’s not about quantity either — the additional cable channels using a digital box are largely in three categories: replicas of channels available in analog cable, additional-fee channels, and music-channels. Of these three, the only ones we ever use are the music channels.
• We feel that a large portion of content created today is not worth watching. Our viewing preferences have actually narrowed somewhat — there are only two channels we watch with any regularity outside the broadcast channels. Yet, our flat fee paid to cable companies does not adequately reward content providers for making the content that we do like.
• More content is available online, or through direct-to-mailbox DVDs from Netflix or Blockbuster. Thus, if we don’t mind waiting a bit for content to become available in either online or DVD format, there’s no need for live broadcast anyway. Even better — when paid for by users, this content is generally commercial-interruption free and better quality than we get through the cable company anyway. I distinguish between commercial-interruption free and commercial free because as we know, the new wave is in product placement on shows. But at least it doesn’t contain those hideously large and non-silent network overlays from channels.

There are others talking this way as well. See this post over at Freedom-to-Tinker for a good read as well. And today, I read that cable companies want to offer exclusive channel content online to subscribers [story]. So this is my plea to programmers. Forget TV stations and network affiliations. Instead, sell your shows direct to viewers. Do it without ads (though I imagine you’ll still have product placement/endorsements), or at least have a two-tiered system where users can pay more for an ad-free program. Then, you will get a better picture of your viewers, and can probably do a better job of marketing to them. Online word-of-mouth can help your show catch on and grab followers. If you are worried about steady-income, offer us high-priced single-show samples, and more reasonable season buy-ins. I would much prefer this — so I can get just the 10-ish shows I actually watch rather than the vast array of TV programming I don’t care about.

So I realized yesterday that all the news about the New Year’s countdowns should be 3-2-1-1 were just dramatizations of reality. I guess I should have figured that all along, but got caught up in it myself. You see, the leap second was added at 23:59:59 UTC (coordinated universal time), and therefore, in my time zone, had been added to the clock 5 hours prior. There was nothing unusual about our countdown (at least, for the last 10 seconds).

But I have yet to find a reporter who actually reported what happened in Trafalgar square (or other places celebrating New Year’s with the UTC leap second in play). Did they start their countdown a second later? Or did they count some number twice? Or were they early? The public wants to know (or at least I do)!