We’re at the end of another month, and I have a bunch of odds and ends. In no real order.
- Never ever forget to back up a .htaccess file because it’s invisible. Either that, or don’t trust that your web host’s online tool to add MIME types won’t copy over your .htaccess file that controls the entire URL structure of your site. Moreover, don’t trust that the host can pull a backup of the file they ruined in less than 24 hours.
- KeepMedia looks really neat. I signed up for the trial.
- Press clipping services are expensive. Just savor the fact that people are talking about you, but you don’t know what they are saying.
- Pair is a group of pros
- If Rich Brome ever decides to make a Mac site, everyone else (read MacX.com) will be toast. PhoneScoop never stops amazing me.
- Yahoo! News is getting better by the month. It’s more useful than Google News for my uses.
- CarbWire has a ton of email list subscribers already (launched the list on Sunday). If you publish for an audience that isn’t tech savvy, don’t count out good ol’ email.
Read all about it. If you don’t like registering, I don’t, use BugMeNot. Great to see people are loving the site. I have added a widget that allows you to sign up for email updates* (check the right sidebar).
Thanks go to Tom Bridge for the software link. I hacked it quite a bit, but it was a solid base.
(On a sidenote, this is the 200th entry to this space. Other stats: 568 comments, the first entry was July 6th, and there have been less than 40,000 hits).
This is exactly why I doubted Feedster would be bought. Not that I would start praising a Microsoft service sight unseen, but the whole business is just too simple. A good search engine (MSN aint bad) combined with a good bot would be great. We’ll have to wait and see on this, but it has potential.
I’m sure Yahoo! will have something similar, especially now that it trolls RSS/Atom for My Yahoo! users. Now if Google would just let me filter out dupliates, I’d be a happy man.
There are a lot of people hating TypeKey before they have even seen it. That’s pretty silly. I’d just like to say thanks to SixApart. They didn’t have to do this, MovableType is a free product. Have some respect for a company that can work for nothing.
Someone said that TypeKey is the PATRIOT Act for weblogging. Besides giving me a good laugh, it gave me one less person I felt the need to read via RSS. There was no Declaration MovableType act that the Senate passed on a rainy night to force us to use MovableType. There’s nothing stopping you from not using TypeKey (it’s only if you want it). So if I don’t have TypeKey, feel free to commit terrorism against my weblog (after all, that’s what the PATRIOT Act stops).
Time to cut the chit-chat, here’s what bugs me about the haters:
1) None of the critics have seen it
2) Said critics feel they have a divine right to post a comment on a given weblog. They don’t. Hell, the biggest critic doesn’t even have comments turned on (Dave Winer).
3) It’s free, optional and open source. What’s to complain about?
Anyone have a suggestion for a good PHP based email list script? Here are my needs:
- Not tag emails as “powered by”. It needs to look like we sent it by hand.
- Work well with a lot of addresses (10-20K)
- Be able to run automatically. I want a daily/weekly newsletter, I want to be able to provide it the content which will self update, and it will send these things out via crontab.
That last one might be a problem, so I might have to write it myself. Any suggestions on how to send large quantities of email? The list won’t be huge for a little while, but I don’t want to hit a brick wall (I assume it will be under 1,000 for a while).
The solution does not have to be free, I’m not objected to paying for quality code.
Another score for Jason Calacanis: Mark Cuban* now was a Weblogs Inc. powered weblog. It’s really good too. He’s making reporters actually report the truth by posting what they say to him before it gets published.
Although, I doubt it’s for money (why would Mark let half the cash go to someone else besides him?), it’s a really smart move by Jason. You just know a bunch of Mav fans are going to swarm the site. I don’t even care about the Mavs, but have subscribed to the RSS feed.
So this is two great moves in a row (and two straight sites that don’t have subdomains or cookie-cutter designs). I like where this is going!
* Mark Cuban sold Broadband.com to Yahoo! which gave him funding to buy the Dallas Mavericks. He’s not liked by many other NBA owners, which makes him 20x cooler.
Thanks to Gawker and The National Post, CarbWire is on its way to being a household name. Fun! And this time I get to be on the news for something I am actually proud of.
Update: being a My Yahoo! daily pick doesn’t hurt either.
I never use Google News to get news on more than a micro-topic, but I have searches saved in-browser and in-reader that provide niche content (e.g. for certain types of cell phones). For that, it works well. But, Google News would be about 3x more handy if it had a -duplicate operator.
When an AP story crosses the wires, every small town news paper and TV station posts it. My once perfect HTML or RSS listing is now full of 10 identical stories from useless sources. And don’t even mention the fact that every single TV station has an identical looking horrid website. Substitute the anchors and it’s complete boilerplate.
How would this operator work? If it’s a Reuters story, have the link go to Retuers. If it’s AP, pick a major national source (USA Today, Yahoo!, MyWay).
Right now on one of my searches, the first page of results is all one story, and from such great sources as:
- Fort Wayne News Sentinel
- Wilmington Morning Star
- Worcester Telegram
- WTOP
Useless. With -duplicate, it should have one link to a reputable national source and then move on with the other stuff. I have to modify my searches to weed out this stuff, or dig through SERPs to find a national source.
I’m looking for a great web traffic stats package. By great, I mean not Analog or Webalizer. Neither give much beyond simple “you had this many hits” numbers.
Here’s what I need:
- Multiple domains – I need to track a network of sites. Ideally, a pull down would let me choose what site I am seeing stats on.
- Archiving – I need to be able to look back 6 months ago and see what the traffic was. Not detailed stuff, but I need to see what was going on.
- Top pages – I need a list of the top pages per given period (e.g. day, week, month, year).
- Proper RSS tracking – Webalizer fumbles when it comes to RSS feeds. I need to see how many people grab the RSS feed. Not how many times it was downloaded, but how many unique times. Plus I need to be able to not count the RSS as a page view.
- Real time referrers – I need to know who’s coming to the site from where now.
- Search engines – I need to track how many people are coming from what search engines.
- Unique per time period – How many unique visitors did we have yesterday? Last week? Last month? In the last 20 minutes?
- OS / Browser
I have a few more “wants”:
- Top incoming searches from X search engine (great for SEO)
- Session lengths
- Average number of sessions per user per month excluding 1 time vistors
- Most popular hour, day of week, and week of month
- OS vs. session length / page views/ avg # sessions
That’s a lot. The ideal system will probably cost $100,000, so I’ll have to chop off a few of the demands. Multi-site / RSS tracking is a must though. I’m tired of looking at 5x inflated values because Webalizer / Analog count the RSS feed as a page.
I have seen 216.187.123.218 request atom.xml on all my sites. Who is it? I assume it’s some kind of Feedster for Atom deal, but it could be anything. Why does it keep asking when the file doesn’t exist? Have you seen it?