Archive for the 'The Web' Category

Jeffrey Zeldman has written an intriguing article on why “maybe” is one option too many. He compares the usage of “maybe” in questionnaires to five-star rating systems, exposing a point I had never thought of. As always, it’s a well written article that makes you think. Zeldman FTW.

Nearly a week ago, a malicious person or group of persons hacked into DreamHost, the company I use to host both Born Geek and this site. The passwords for over 3,500 FTP accounts were compromised, and some customers found unauthorized changes to files or directories. My account was among those that got hacked, and the experience has made me a better computer user.

Having a password stolen is frightening enough, but my situation was nearly a worst case scenario. When I originally set up my user account with DreamHost, I naturally provided the password to be used with that account. This user account granted me access to the DreamHost web panel, FTP uploads, and access to the web server’s shell (via either telnet or ssh). When I later set up an email account, I chose to use this same account out of sheer convenience. I made a likewise decision for access to my web server logs. So, in short, one username and password provided me access to five areas:

1. The DreamHost Web Panel
2. My web storage (via FTP)
3. My web server home directory (via telnet or ssh)
4. My primary email address
5. My server logs

Do you see the problem here?

As soon as I got the email that my FTP password had been compromised, I realized how slack I had been about security and panicked. Thankfully, none of my files or databases were corrupted (though I’m still taking a look through everything). I have since changed all of my passwords, and they now all differ from one another, something I should have done from day one.

I try to be as security conscious as possible, but I really dropped the ball in this area, mostly for convenience’s sake. This is the first time I have been ‘hacked’ like this, and I’m actually glad it happened. The experience has motivated me to be more secure in my password handling.

Lots of people are jumping ship as a result of this, but doing so seems premature to me. The folks at DreamHost are being open and honest about the problem, and I really appreciate that. Any company that steps up and says “we made a mistake and we’re trying to prevent it from happening again” is worth sticking with. At the very least, I’ve learned a much needed lesson.

Update: I forgot to mention that other web hosts were also hit by this attack (according to this post), so it wasn’t solely a DreamHost issue.

I have fixed a bug in the navigation links in MonkeyAlbum here at the site. Some of the links used to navigate photo albums with multiple pages were pointing to the wrong location. If you happen to spot any other problems, please let me know.

A new photo album (possibly my best one yet) will be posted within the next day or two, so stay tuned.

I am growing increasingly frustrated with Google Maps. In the past month, on two separate occasions, Google Maps failed to find my intended destination. What really gets under my skin is the fact that Google’s competition found each place without any problems.

Example 1
My family checked out the Clarksville Station restaurant in Roxboro, NC for my sister’s graduation. It’s a steak-house built inside of an old train station and a couple of dining cars from an actual train. It’s located at 4080 Durham Road, Roxboro, NC. Let’s see what the mapping services show for this query:

• Google Maps: Only locates Durham Road, not the 4080 address.
• MapQuest: Shows the location as expected.
• Yahoo! Maps: Can’t find the exact location, but interestingly enough, centers the map at the exact location. Weird.
• Microsoft Live Search: Gets it exactly right.

Example 2
My car needed service recently, so I took it to Jay’s Automotive, a repair place not too far away from where I live. They are located at 3510 Highway 70 West, Efland, NC. Let’s see how the various mapping services do with this one:

• Google Maps: Wow. This is so far off, it’s not funny.
• MapQuest: Again, MapQuest gets the location exactly.
• Yahoo! Maps: Again, they cannot locate the address, but the map is centered at the correct location.
• Microsoft Live Search: Again, Microsoft got it exactly right.

What gets me even more steamed is the lack of aerial (or satellite) images for example number 2. Google Maps only has images beginning at zoom level 6 (levels 1 through 5 are all “unavailable”). MapQuest has color images down to zoom level 3 (1 and 2 aren’t available), which is very close. Yahoo! Maps has color imagery at all zoom levels, while Microsoft Live Search has images to zoom level 3 (just like MapQuest, though the image quality is very poor).

Google needs to stop spending money and effort on cheap gimmicks like their recent Street View (is that thing worthless, or what?), and instead beef up their location database and aerial photographs. I can’t even see my house on Google Maps! All the other mapping services have it, so it’s certainly possible to do.

Get with the program, Google. Until then, I think I’ll stick to your competition (at least when satellite photographs are involved).

Last Thursday, Google unveiled their new universal search, complete with a change to the look and feel of their website. Now when a user searches Google, not only are web results returned, but other search results (e.g., news and product results) are also returned. What’s most convenient is the fact that these additional items are inserted quietly amongst the web results, not directly above or below them. This new search paradigm is interesting, and it will be interesting to see if it holds on. I particularly like the subtle changes Google made to their look and feel, with that slick little menu bar running across the top of all pages. Having direct access to all the various search types is very handy.

Matt Cutts has posted a short collection of improvements he’d like to see over at Amazon.com. I agree with him on all counts. There are a number of areas that Amazon could do way better on; hopefully some of these ideas will see the light of day.

Judging by the comments on the post, it also looks like I’m not the only Amazon Prime junkie.

An article over at Slashdot mentions the news that Apple, Mozilla, and Opera have requested the adoption of HTML 5, a successor to HTML 4 and XHTML, both of which are aging. The draft spec for HTML 5 is courtesy of the WHATWG, a group that formed for the very specific purpose of creating a successor to what we use today. It seems that the World Wide Web Consortium (W3C) has been wasting time working on an incompatible replacement to HTML and XHTML, a decision that I feel is destined to fail (backwards compatibility is a definite requirement in moving the web forward).

Adopting HTML 5 is a great idea (HTML does need cleaning up), but unless Microsoft climbs aboard, we might as well start building the HTML 5 coffin. As much as people dislike hearing it, the fact is that Microsoft owns this kind of thing. If it doesn’t get into Internet Explorer, you might as well forget it. And I doubt that IE will migrate to something like this anytime soon. I mean, they just fixed their CSS box-model bugs! How many years did we wait for that? And after all that time, the rendering engine is still broken.

Perhaps Firefox will continue to take share from IE, giving the ‘little’ guys like Mozilla a voice in what direction to take the web. I’ve certainly got my fingers crossed.