Stories
Slash Boxes
Comments

News for nerds, stuff that matters

Slashdot Log In

Log In

[ Create a new account ]

sootman (158191)

sootman
  (email not shown publicly)

Journal of sootman (158191)

Sp, what is there to do in D.C.?

Friday May 16, @01:07AM
User Journal
I'm going to Wasington, D.C. next month for the DC PHP conference. I've been there once before with family and enjoyed it a lot but this time I'm going alone and I'm wondering what there is to do in town in the evening. Seems like all the museums close around 7 or 7:30 and I don't remember there being much else to do at night but walk around looking at the monuments. I'm not planning to rent a car while I'm there. Is there anything to do there? Any movie theaters even?

Changes in 'screencapture' from Tiger to Leopard

Monday April 28, @03:36PM
OS X
OS X ships with a great little command-line utility called 'screencapture.' It is a command-line interface to OS X's screen capture tool, which is activated from the desktop by pressing shift-command-3. If you want to monitor a computer you're not sitting at (10.4 or earlier--see below) just ssh in and say something like

screencapture -x screen.png
and then view the file--scp it to the computer you're sitting at, or use FTP, or whatever. It's a standard unix-style command, simple as can be--'screencapture' does just what you'd expect it to, the '-x' means "don't play the 'click' sound" and then you specify the output file, 'screen.png' in this example. (This worked in earlier versions of OS X with some changes. 10.3, for example, saved screen caps as PDFs so you'd say 'screen.pdf' instead.)

You have to run that as the logged-in user or as root. One of the things I used it for was to monitor some conference rooms that I'm in charge of by running this script and saving the file to a web-viewable directory. To do this on a stock 10.4 machine, just create a file called 'screen.pl' in /Library/WebServer/CGI-Executables/ with these contents:

#!/usr/bin/perl
use CGI;
$cgi = CGI->new();
print $cgi->header;
print qq~<html>\n<head>\n<title>View Screen</title>\n</head>\n<body>\n~;
my $step1 = qx!screencapture -x /Library/WebServer/Documents/screen.png!;
print qq~<img src='/screen.png' border='1'>\n~;
print qq~</body>\n</html>\n~;
1;
It's a Perl script, so 'chmod 755' it, and there should be a blank line at the end, IIRC. Make sure the user this script will be running as can write to /Library/WebServer/Documents/. Then, edit your httpd.conf file (sudo pico /etc/httpd/httpd.conf) and change the user that the server runs as to either a) the user who is always logged in or b) root. (Yes, I know this is a potential security risk. I do not recommend running this on a box that is viewable to the Internet at large.) In my case, monitoring computers in the conference room where everyone logged into an account named 'conference', I changed these lines in httpd.conf from

User www
Group www
to

User conference
Group conference
(Remember to restart the webserver for these changes to take effect. (sudo apachectl graceful) Also note that dpending on various things, the group name may or may not be the same as the user name. ls /Users/ for a hint.) Once you've done that, you can see the screen by visiting http://ip.address.goes.here/cgi-bin/screen.pl . Very useful little tool.

So everything was going along fine until 10.5 came out. They seem to have slammed the door on running screencapture in any way remote way. Even if you ssh in as the user who is currently logged in it won't let you. But you can then log in locally, press 'up' one time to call the same command out of .bash_history, and it works just fine. And, of course, my perl script above no longer works. (Note that 10.5 has Apache's config file stored in /etc/apache2/httpd.conf.) Well, it kind of works, but all you see is a black image (the same size as your desktop) and not the actual screen. But it's definitely a remote-vs-local issue: you can run say screencapture -x screen.pngfrom a locally-spawned Terminal session and it works--creates a file in whatever directory you're in--but ssh in and run it or visit /cgi-bin/screen.pl and it doesn't. The question is, does anyone know why, and can anyone find a way around this? I've tried some things with Automator but I really want this to run invisibly so it doesn't bother whoever's using the machine.

How to make Slashdot better

Friday January 18, @02:55PM
Slashback
These thoughts are by no means original. I just have them rattling around in my head and putting them down onto (virtual) paper will free up resources for other tasks. I'll write this in terms of "you" rather than "they" in the hopes that someday, someone who actually has input in how Slashdot works will read these ideas and implement them. (Update: OK, looking this over, I see I kind of meandered around and change person and tense throughout. Maybe I'll fix that someday.)

Slashdot is great. But there are a few rough edges that continually poke through the surface, like springs in a couch, that spoil the experience to various degrees. It would take so little effort to make these changes and the rewards would be so great--truly an example of the 90/10 rule.

Let's start with the two biggest gripes about Slashdot: things that come up so frequently and have been around for so long that they're practically clichés: dupes and grammar.
  1. It is absolutely trivial to check for dupes manually. Just use Google (Slashdot's own search isn't that great) to see if the topic has been covered before. If it seems to have been covered before, look closely to see if this is the same news or something related or a follow-up. If it is, feel free to post the new story, but make it clear that this is new and not a dupe.

    It is almost as trivial to create an automated system to check for dupes. Just have something that looks for keywords in the subjects, checks the URLs in the story, etc. If there is a match, raise some flags.
  2. Grammar: please. You call yourselves 'editors' but you are, at the moment, wholly undeserving of that title. Start with a spellcheck. Proceed to a grammar check. Here's an idea: have someone else look it over, too. I don't expect everyone on the staff to have a Master's in English. But there are plenty of typos, misspellings, and other major errors, to say nothing of awkward phrasing and other unpleasantness, that trip people up when reading the summaries. You are editors, not journalists: your job is to edit submissions, making them worthy of reading, not to treat the submitter as a source whose words you have a sacred duty to present exactly as delivered. Take the time to make it good. This isn't a race. You're not "scooping" anyone. This isn't Digg.

    And the same goes for the headlines. Headline writing, like everything else, is an art. Plenty of Slashdot headlines have to be read more than once to make sense. On the one hand, yes, they should by nature be brief; on the other hand, this is not a newspaper and every inch is not infinitely precious. Go for clarity over brevity.

    And for God's sake, quit being downright inaccurate or misleading with the headlines. I mean that's just common sense. There is bias, there are slants, there are interpretations, but being downright wrong is inexcusable.

    For tips on grammar, news writing, and headline writing, go here and here.
There are three other major things that need to be fixed with submissions.
  1. Don't link to blogs that link to the story. Yes, the submitter supplied the story, but ten other people probably did, too. They can supply a name, if if they supply a URL their name gets made into a link as well. That is reward enough. If he's hoping for page views and ad impressions, that will have to do. There is no reason I should look at his two-line "Check out this story on Ars/Anandtech/news.com/NYT" blog entry on my way to get to the actual story in question.
  2. Don't editorialize too much in the submission. And for the love of God, quit adding dopey questions to the end of submissions: anything from "What do you think?" to that hated old standby, "Is this the end of Microsoft/Windows/Office/Linux/Apple/SCO/America?" This is a discussion site. People who come here know this. We don't need a third-grade-style prompt to get the discussion going. All that does is generate two hundred posts of "No, Zonk, you're an idiot" (along with the inevitable and useless "yes/no/maybe" tags) and take away from the real discussion.
  3. This is an issue less than some other things but it is absolutely essential: do some fact checking. Check to see if this is five-year-old news that some random guy just now discovered. Check to see if it's a hoax, or an out-of-date item that has since been debunked/refuted/disproved. Read the article and see if the submission is even correct. Here's on example of posting a complete mistruth. Thankfully this was corrected, but that's a relatively rare case.
There are plenty of minor things as well.
  1. Be aware that not everyone knows everything. This site does have a certain audience with some things in common, so we don't need a link to clarify what Linux is, but many other things could use some clarification, especially new and/or uncommon acronyms. (Speaking of acronyms, follow the old rule: say what they mean the first time they appear. Just because it didn't occur to the submitter to spell it out until he had written it three times, doesn't mean you have to do the same. Remember: edit!) Use your judgment, and show it to at least two other people. Again, not doing so devalues the discussion: rather than talking about the topic, you get 50 posts telling you what the topic is.
  2. Read the comments. Update the story as needed. If a bunch of people write to say that a story is wrong, fix it! This is the Internet, not a print piece. Make the change--with an 'Update:' notification, of course--but for God sakes, don't let errors of fact sit there forever. I don't expect the editors to be perfect in their application of the above tips, but I do expect them to take advantage of the millions of volunteer helpers who point out errors.
Other:
  1. Don't allow ACs to post links.

Apple Prediction Part II

Thursday July 12 2007, @01:30PM
OS X
Weak! My previous journal entry, which I planned to edit by adding comments, has been archived, so comments can't be added. Oh well. I'll just put commentary into this one. Original = italics, new = plain text. Old text has been trimmed. Read the original if you want.

- The iPhone will do quite well. There WILL be lines at 6pm on 6/29. (I'm going to the mall that day to see for myself and will post links to pics/vids in the comments below.)

Well, another commitment prevented me from making it out, but there were lines. (I made this point because someone, somewhere--not just a nobody, but someone you might have read; I wish I remembered who--said the iPhone hype was fake and there wouldn't be lines.) Not only were there lines, but they were shown on most major news channels. How well did it do? A bit of searching puts it at 500-700 thousand units in the first weekend. Bloomber.com says "The iPhone broke AT&T's opening-weekend records, selling more in three days than phones such as Motorola Inc.'s Razr did in their first month, according to spokesman Michael Coe."

- Once the iPhone is out, we'll see some more hardware updates, probably starting in August or September. So far this year all we've seen are slightly faster MacBooks, a move to Santa Rosa in the MacBook Pro, and the addition of an eight-way Mac Pro. Except for the 8-way, the Mac Pro line hasn't changed since its introduction almost a year ago. The Mac Mini is also very overdue for an update--it is overpriced and underpowered, and since I've got a G4 Mini that I want to upgrade--but won't at the current price--I hope Apple doesn't let this line languish.

I'm 2 for 2 so far (or a bit more; see below) and wow, when it rains, it pours. Today (August 7) we saw...
- the new iMac
- upgrades to the Mac mini (Core 2 Duo CPUs)
- upgrade for the Airport Extreme Base Station (gigabit ethernet)
- and, something I didn't expect to happen--upgrades to iLife, iWork, and .Mac.
(When making my predictions, I didn't say anything about Apple's software products.)

There is also some new software for the iPhone (in addition to last week's update) as I mentioned below. To be honest, I'm looking forward to much larger updates (video recording!!!) but I'm surprised, pleased, and impressed by how many updates Apple has dropped for the iPhone and how quickly they're coming.

- There will NOT be a touchscreen iPod until after Christmas AT THE EARLIEST. Yes, it's a nice technology. Yes, Apple can and will release one eventually--hell, they could have released one ALREADY if they wanted to, right? I mean, what's easier to make: and iPod or an iPod/phone/Internet thingie? DUH. They KNOW they can. They KNOW people want them. But they DON'T want them to cut into iPhone sales. I bet they won't even release 100/120 GB iPods until then, or change the Shuffle or Nano lines. January at the ABSOLUTE EARLIEST; maybe in time for Valentine's day (like they did with the introduction of the colored Shuffle this year), otherwise wait for Spring or Summer.

- Similarly, the iPhone WON'T get minor revs (higher capacity, lower price) until after Christmas. Major revs, like the ability to use faster-than-EDGE cellular data networks, won't arrive for 9-15 months. Some improvements could be handled just in software updates--the ability to use native IM, or send MMS, or record video with the built-in camera--but given the recent debacle over the $1.99 802.11N update, I'm not sure which of these Apple could/would roll out for free. If they don't come free in a software update, look for them in iPhone 2.0, Spring/Summer 2008. Faster networking WILL be in the next major rev of the iPhone.


Wow, was I ever wrong here. Turns out Apple does not want to milk huge margins out of a handful of customers, and force everyone who wants a touchscreen iPod to buy an iPhone. Turns out they'd rather make aggressive moves and try to capture as much market as possible. The special event on September 5 brought out an ENTIRELY new iPod line, ranging from new colors on the Shuffle to video and a new shape for the Mini to a huge 160 GB capacity for the Classic plus the introduction of the Touch plus a staggering $200 price drop on the iPhone. Wow. I'm speechless. One note regarding margins--if they were able to drop the price $200, that pretty much means the profit on each (original) iPhone was over $200. Wow. Nice. Let's just say the profits went from $250 down to $50--that means they've got to sell 5x as many iPhones as they originally planned to make the same amount of money. We'll see how they do. Overall, this is a big move to solidify the iPod's market dominance.

- As I've said, I have no plans to get an iPhone any time soon. I predict they'll be $249 by mid-2009, at which time I'll buy. Unless my current cell phone dies--then I *might* buy one. But probably not even then. Long story short: as cool as the iPhone is, it doesn't do anything I *need*, or even really *want,* outside of the usual hey-that's-a-neat-feature. (Google maps! Cool! Too bad I don't ever go anywhere but back and forth between my house and my two jobs.) I like my current phone, and it's smaller than the iPhone is or ever will be, and *that's* really important to me. (I don't carry a PDA, because I hate carrying big, heavy crap, and my iPod only gets used in my car.) The only thing I'd like would be a better (~1MP) camera--mine only does 320x240. The introduction of true GPS might also push me to buy one--that's one device I *do* want (I've got one, but with a tiny B/W screen) and would be willing to carry.

And finally, I got this one wrong too. :-) I got an unexpected raise at work the same time the price came down, so I got one of the last (and cheapest) 4 GB models. (OTOH, I got it at the price I wanted--$249.) Overall, I'm happy enough. For phone-related stuff, my old one did some things better. The iPhone obviously does more, and it's quite cool and fun to play with, but it's not better in every single respect. I felt a bit sad when I switched my account over and my old one complained "SIM card not recognized" or whatever the message was.

Mid-2007 Apple predictions

Wednesday June 27 2007, @06:10PM
OS X
Here are my Apple predictions.

- The iPhone will do quite well. There WILL be lines at 6pm on 6/29. (I'm going to the mall that day to see for myself and will post links to pics/vids in the comments below.*)

- Once the iPhone is out, we'll see some more hardware updates, probably starting in August or September. So far this year all we've seen are slightly faster MacBooks, a move to Santa Rosa in the MacBook Pro, and the addition of an eight-way Mac Pro. Except for the 8-way, the Mac Pro line hasn't changed since its introduction almost a year ago. The Mac Mini is also very overdue for an update--it is overpriced and underpowered, and since I've got a G4 Mini that I want to upgrade--but won't at the current price--I hope Apple doesn't let this line languish.

- There will NOT be a touchscreen iPod until after Christmas AT THE EARLIEST.** Yes, it's a nice technology. Yes, Apple can and will release one eventually--hell, they could have released one ALREADY if they wanted to, right? I mean, what's easier to make: and iPod or an iPod/phone/Internet thingie? DUH. They KNOW they can. They KNOW people want them. But they DON'T want them to cut into iPhone sales. I bet they won't even release 100/120 GB iPods until then, or change the Shuffle or Nano lines. January at the ABSOLUTE EARLIEST; maybe in time for Valentine's day (like they did with the introduction of the colored Shuffle this year), otherwise wait for Spring or Summer.

- Similarly, the iPhone WON'T get minor revs (higher capacity, lower price) until after Christmas. Major revs, like the ability to use faster-than-EDGE cellular data networks, won't arrive for 9-15 months. Some improvements could be handled just in software updates--the ability to use native IM, or send MMS***, or record video with the built-in camera****--but given the recent debacle over the $1.99 802.11N update, I'm not sure which of these Apple could/would roll out for free. If they don't come free in a software update, look for them in iPhone 2.0, Spring/Summer 2008. Faster networking WILL be in the next major rev of the iPhone.

- As I've said, I have no plans to get an iPhone any time soon. I predict they'll be $249 by mid-2009, at which time I'll buy. Unless my current cell phone dies--then I *might* buy one. But probably not even then. Long story short: as cool as the iPhone is, it doesn't do anything I *need*, or even really *want,* outside of the usual hey-that's-a-neat-feature. (Google maps! Cool! Too bad I don't ever go anywhere but back and forth between my house and my two jobs.) I like my current phone, and it's smaller than the iPhone is or ever will be, and *that's* really important to me. (I don't carry a PDA, because I hate carrying big, heavy crap, and my iPod only gets used in my car.) The only thing I'd like would be a better (~1MP) camera--mine only does 320x240. The introduction of true GPS might also push me to buy one--that's one device I *do* want (I've got one, but with a tiny B/W screen) and would be willing to carry.

* I plan to NOT edit this document so the time/date stamp won't change. Any updates will be in the comments below.
** excuse the caps. Too lazy to add bold/ital tags. :-)
*** For those that don't know, the iPhone can do SMS, and it can send pics with email, but it can't send pics with SMS (pics+SMS=MMS)
**** too bad it can't do video. With the YouTube partnership, I'd think that the ability to shoot video and upload it to YouTube directly from the iPhone would be a great feature.