The last few days were interesting, to say the least.
I don't exactly know what's going on in the irony department up there, but it feels like someone's determined to make me waste my time. Apart from the usual real-life issues (including but not exclusively, my driving license, of which you can see a picture on my Facebook page because, yes, I'm that shallow), and from the PC apocalypse I told you about in my last post, I also faced two new problems this week-end.
First of all -- the new computer I bought has the Intel P67 bug with the SATA interface. Meaning half of my SATA ports are not usable. One of them was being used by a hard drive so I just moved it to a safe port, but it now means all 4 of my safe ports are taken (plus my only eSata port), and I can't add any other hard drive. Thankfully I didn't exactly plan to do that either, as I already have USB3 ports which will come in handy in the future.
The only question then, is whether or not I should bring back my computer to the shop. I mean they're not supposed to sell me a buggy motherboard when Intel recognized the bug and everyone recalled their motherboards months ago already. I'm definitely unlucky these days.
Also in the let's have fun category, my ISP apparently decided to go play with my Interwebs cables, and overnight my download rate went from about 4 Mbps to under 10 Kbps. Yes, I'm talking about bits per seconds, not bytes. See, I've been on the Internet since March 1996, at which time I had a trusted 28.8 Kbps US Robotics modem. Anyone remember these? They were the top modem makers at the time. But they still ran at 3 kilobytes a second. It wasn't funny. Well, my bandwidth was one third of that for nearly 48 hours. Back in '96, web pages were just a few kilobytes of HTML, possibly a few images, later a CSS file or two, but jQuery couldn't have even existed at the time. Even the first versions, which were very light and quick to load, would take many seconds to load on an old modem. This is one of the reason why I spent so much time working on reducing the footprint of both jQuery and general JavaScript code in Wedge, when I could just have stopped worrying and dropped the compressor. I was adamant that Wedge SHOULD be faster than SMF in every possible way, and most of the work had to be done on the final output. It took me weeks just to get to the point where, comparing stock copies of Wedge and SMF, I concluded that Wedge felt faster to load than SMF. It's a personal feeling, see, it can hardly be calculated with a stopwatch. But even reaching that point was hopeless to begin with, because we'd decided on including jQuery. So we got the best of both worlds. But I still have to fight myself to determine what version of jQuery we'll ship with in the end. I mean, we started Wedge with 1.4.4, then switched to 1.5.2 with absolutely no benefit to us, and then to 1.6 where I reported a handful of annoying bugs, and once again I don't see any benefits... I guess I'm waiting for 1.7 to determine whether it's worth upgrading. But right now, as it is, I'm actually tempted to go back to 1.4.4, and save the extra 5-6 KB that 1.6.1 uses. 5 to 6 KB may not be much to you, but it's worth a lot to others.
And somehow, after my horrible Sunday experience, I can't help but wonder: won't somebody please think of the analog modems? These days, it's all about working from home and avoid commuting. So, people can go live anywhere they want. What happens to those who happen to find a heavenly place with no DSL lines around? What happens when all of the web is optimized for broadband? Will analog users end up as the digital man's stoneman? Be viewed as disabled, maybe? WAI-ARIA to the rescue, or no future for losers?
We need to stop wasting resources. Just because a font looks cool doesn't mean you should force your user's browser to download it every time they come to your website. Interestingly, that's exactly what I did a few weeks ago. I went for PT Sans as the default header-type font in Wedge, but decided against specifically calling it from Google Fonts, as fast as the service is. I just don't see the point in doing that when you can simply tell your users to download the font and install them by themselves. Then they no longer have to download it again -- and the font can start popping up everywhere.
Install a font and save a tree!
Find your raison d'être!
To arms, my brothers!
I don't exactly know what's going on in the irony department up there, but it feels like someone's determined to make me waste my time. Apart from the usual real-life issues (including but not exclusively, my driving license, of which you can see a picture on my Facebook page because, yes, I'm that shallow), and from the PC apocalypse I told you about in my last post, I also faced two new problems this week-end.
First of all -- the new computer I bought has the Intel P67 bug with the SATA interface. Meaning half of my SATA ports are not usable. One of them was being used by a hard drive so I just moved it to a safe port, but it now means all 4 of my safe ports are taken (plus my only eSata port), and I can't add any other hard drive. Thankfully I didn't exactly plan to do that either, as I already have USB3 ports which will come in handy in the future.
The only question then, is whether or not I should bring back my computer to the shop. I mean they're not supposed to sell me a buggy motherboard when Intel recognized the bug and everyone recalled their motherboards months ago already. I'm definitely unlucky these days.
Also in the let's have fun category, my ISP apparently decided to go play with my Interwebs cables, and overnight my download rate went from about 4 Mbps to under 10 Kbps. Yes, I'm talking about bits per seconds, not bytes. See, I've been on the Internet since March 1996, at which time I had a trusted 28.8 Kbps US Robotics modem. Anyone remember these? They were the top modem makers at the time. But they still ran at 3 kilobytes a second. It wasn't funny. Well, my bandwidth was one third of that for nearly 48 hours. Back in '96, web pages were just a few kilobytes of HTML, possibly a few images, later a CSS file or two, but jQuery couldn't have even existed at the time. Even the first versions, which were very light and quick to load, would take many seconds to load on an old modem. This is one of the reason why I spent so much time working on reducing the footprint of both jQuery and general JavaScript code in Wedge, when I could just have stopped worrying and dropped the compressor. I was adamant that Wedge SHOULD be faster than SMF in every possible way, and most of the work had to be done on the final output. It took me weeks just to get to the point where, comparing stock copies of Wedge and SMF, I concluded that Wedge felt faster to load than SMF. It's a personal feeling, see, it can hardly be calculated with a stopwatch. But even reaching that point was hopeless to begin with, because we'd decided on including jQuery. So we got the best of both worlds. But I still have to fight myself to determine what version of jQuery we'll ship with in the end. I mean, we started Wedge with 1.4.4, then switched to 1.5.2 with absolutely no benefit to us, and then to 1.6 where I reported a handful of annoying bugs, and once again I don't see any benefits... I guess I'm waiting for 1.7 to determine whether it's worth upgrading. But right now, as it is, I'm actually tempted to go back to 1.4.4, and save the extra 5-6 KB that 1.6.1 uses. 5 to 6 KB may not be much to you, but it's worth a lot to others.
And somehow, after my horrible Sunday experience, I can't help but wonder: won't somebody please think of the analog modems? These days, it's all about working from home and avoid commuting. So, people can go live anywhere they want. What happens to those who happen to find a heavenly place with no DSL lines around? What happens when all of the web is optimized for broadband? Will analog users end up as the digital man's stoneman? Be viewed as disabled, maybe? WAI-ARIA to the rescue, or no future for losers?
We need to stop wasting resources. Just because a font looks cool doesn't mean you should force your user's browser to download it every time they come to your website. Interestingly, that's exactly what I did a few weeks ago. I went for PT Sans as the default header-type font in Wedge, but decided against specifically calling it from Google Fonts, as fast as the service is. I just don't see the point in doing that when you can simply tell your users to download the font and install them by themselves. Then they no longer have to download it again -- and the font can start popping up everywhere.
Install a font and save a tree!
Find your raison d'être!
To arms, my brothers!
