No. Unless You Pay Us Lots Of Money.

I received the official word today from the “powers that be” at that they will not adjust their Reaper settings to accommodate Movable Type and publishing via XML-RPC. I understand their desire to keep their servers safe and functional. It’s good business practice. The thing that is a bit annoying is that I’m asking for support for something that is very common on the web. Commercial software that thousands of people use.

What made me angry was the suggestion they made that I should get a dedicated server because then I wouldn’t have any restrictions. Anyone can run MT with a MySQL database at for $18/month. I pay a little more for additional databases and more space and some other goodies. But their “solution” is for me to pay $150+ a month. That’s ridiculous and insulting.

I’m trying to calm myself and remind myself that they are not bad people. They really are just trying to keep their servers running at peak efficiency. And that telling someone that they can make a perfectly reasonable request happen with a 500% increase in fees is a simple error in judgment on their part. They should have just left it at we’re not changing our policy at this time.

(Of course all of this is possibly moot since I have successfully posted via ecto several times since my site redesign over the weekend. Let’s see how this one goes….)

I’m Shocked

Both my previous post and an update I did to it went through properly to my server. The Reaper didn’t kill either. I don’t dare hope that either has changed things to allow XML-RPC to function properly or MT3.1 changed things enough that it’s no longer a problem. More likely this is more of the same thing I’ve been seeing for the last several weeks. Sometimes the posting works. Sometimes it doesn’t.

I might let myself have a little bit of hope though. Ecto really is a great piece of software and I really would like to use it.

Movable Type 3.1 & Ecto 2

I upgraded to MT3.1 last night. I haven’t really done anything with it other than check out all the different interface pages. Since I had already been using MT3 for several months, 3.1 did not seem like that radical of an update. Obviously its support for dynamic pages is pretty huge but since as far as the interface went, that was relegated to 3 radio buttons on the Templates page it didn’t appear significant.

The subcategories looks pretty cool too. Very simple interface. As much as it sometimes pains me, I don’t use multiple categories, I don’t think I’ll get into the whole subcategory thing. At least not with this blog. It could definitely be helpful, I wouldn’t rule it out for the future.

I’m typing this up on yet another ecto 2 beta. I don’t have much confidence that it will get properly published. I think that The Reaper will probably kill it. And since I can’t reliably use it with my host, I probably won’t use it at all. It’s really too bad because Adriaan really does have a nice program.

I had never used ecto before going into this beta. I didn’t really come with any expectations—though I had read lots of people mention how much like liked it, so I figured it would be pretty good. And it really is. For me the two best things are the database and the spell check. Maybe I’m not very demanding with my publishing needs, but it’s true. I love the fact that I can search out old entries and paste the URLs to them into newer posts. And the fact that ecto points out my spelling errors immediately after I type them is a god-send. Everything else is icing.

I don’t need it to do a ton of things. Probably the next biggest thing that I do that’s a bit of a pain is deal with images. That’s something that on one hand ecto handles nicely, but on the other I don’t feel that it goes far enough. After you drag and drop a picture into your editing window, you can click the link that is makes to change how it works.

There are a lot of great options: It’ll make thumbnails of large pictures for you. You can set the directory it should be uploaded to on your server. There’s even a template where you can exactly format the code that will be generated. This is what I would like to have changed. I love digging into advanced features but I also want things to be quick and easy whenever possible. It would be great if ecto would allow you to save various code templates in the image window and access them like a favorites or presets pulldown menu in many programs. So in my perfect world of ecto, I might have one favorite image template that is setup with HTML4 code for one particular blog that I might write for. Another template for XHTML and a third with Markdown code. Since the idea behind ecto is that you can write for and publish to all your weblogs, it should support all the different format possibilities at once. Even with those basic code differences, it would be nice to have one template for code when a thumbnail is generated and another for when the image is already the proper size.

In this way I imagine that embedding pictures into your posts with ecto would involve a drag and drop, a click to bring up the image menu, and then a selection from a favorites menu and click ok. Fast, simple, easy. That’s all. Sure you might have to make a couple adjustments once in a while, or you might want to type something more extensive in the “alt” field but it should be really quick. Don’t get me wrong, as it exists the image window is very easy to use and you can set it up one time for HTML4 and the next time for XHTML but there’s a bit of typing and clicking involved in that. I’d just like to see it taken to the next level of functionality.

After that my list of changes get into nit-picky things. When you’re working in an editing window, there’s no immediate indication of which weblog you’re typing an entry for. Ecto offers support for multiple blogs on multiple servers. Each server is known as an account. There’s a pulldown menu immediately below the “Title” and “Keywords” field which shows the account, but not the individual weblog on that server. So in my case it shows “” but it’s not until I click on that menu and drag over to the sub-menu that I see “Monsters from the Id”. Minor point, but again, since it’s supporting multiple weblogs, it would be a good idea to make things very clear and easy for everyone.

There are a lot of great things which I think many users would find extremely helpful. You can copy your Movable Type template (or a simplified version of it) into the “Preview Template” window. That in conjunction with ecto grabbing your CSS file from your website allow you to have a decent idea of what your text will look like while you’re typing. Adriaan has setup a tags menu with all the most important HTML tags and it’s fully editable so you can add your own tags and assign hot keys. This is actually another area I thought he should modify a bit. Again with the idea that different weblogs might be done different ways, I suggested supporting either different sets of tags menus so that when you’re working in an HTML4 weblog it brings up a set of HTML4 tags versus an XHTML weblog with XHTML tags. Or by adding in menu dividers, you could group appropriate tags together. So for example, the first set of tags is XTHML and after the divider they’re all Markdown.

If it weren’t for the fact that my host kills off the MT XML-RPC script for consuming too many resources on the server, I would definitely be using ecto. Though I would definitely continue to bug Adriaan to make the changes I suggested. 😉 Since I can’t post reliably, I probably won’t. And since the new BBEdit 8 has a much improved spell check, one of my two major needs is satisfied right there. I’ll probably stick with BBEdit until my host lets the XML-RPC scripts run properly or until I switch hosts—neither of which will happen anytime soon.

Don’t Fear The Reaper

The good news: Thanks to a lot of patience from the wonderful Adriaan and some great advice from the fantastic Ben Trott, we’ve figured out why I can’t always successfully post to my website from ecto.

The bad news: I don’t know if we’ll be able to fix it. has a program called The Reaper. It’s purpose is to protect their servers from runaway programs by killing off ones that use up too many resources. (Perhaps they should call it Tom Selleck.) One of the criteria is that a program cannot use more than 16MB of memory. MT-xmlrpc.cgi, the perl script that handles remote posting for Movable Type, is often hitting that 16MB barrier and getting killed by The Reaper.

I have held a lengthy email discussion with various Pair support staff about this with a final plea to look into increasing this memory limit. They assured me that my request was passed on to the appropriate people for consideration.

What would be great is if other users would step up and say “Hey, I think it would rock if the memory limit was increased to 24MB!” Or even better since I don’t really know how much memory is need for XML-RPC, “Hey, I would be swell if we had enough memory to run MT, including XML-RPC, without getting a visit from The Reaper.”

Perhaps This Will Fix My Problems

I continue to have posting problems with ecto. I started searching the Movable Type forums to see if anyone has any ideas. I’m trying something out here to see if it helps.


Nope. Several months ago, people reported problems with using ecto and Markdown. They were able to fix things by submitting with the standard “Convert Line Breaks”. I tried it, and it didn’t fix my problem. Going back to Markdown.


Trying something else… nope.

Request For Help: MT + + ecto

I would love to hear from anybody who has a Movable Type website hosted by and is using ecto to publish. I’ve been attempting to beta test the new version 2, but I’ve been having problems simply posting. Adriaan has said it’s a rare thing that he’s seen twice before where certain combinations of software don’t work well with ecto. So if anyone is doing this please email or post a comment below. I’d be interested to find out about the version of MT, the version of ecto, whether you’re using a cgi wrapper, what settings you had to change in your mt.cfg file, and any other tips or tricks you might have.

Thanks in advance for any insight you can offer.


This is the information about my setup. I am hosted by on a Pentium 4 running FreeBSD 4.8-STABLE, Apache 1.3.29, Perl 5.8.3, and MySQL 4.0.14. I have Movable Type 3.01D installed and running as a standard cgi process accessing a MySQL database. The necessary directories have their permissions set to 777. And in my mt.cfg file, I have the line NoTempFiles 1 uncommented. I’ve been using this setup since the beginning of March with no problems. (Well, actually back then it was MT2.661 and up until a few days ago, I was running under a cgi wrapper with permissions set to 755.) Posting from MT’s web interface has always worked great for me. The worst thing I’ve encountered is sites I’ve tried to ping timing out.

Now I’ve been trying to use the various beta versions of ecto 2 without much success. When I was running under the cgi wrapper, I couldn’t get the posts to go through at all. Or at least the server wouldn’t acknowledge them. I’d get the error I mentioned in this post. The posts would usually appear on my website but without a category.

Since I’ve turned off the wrapper, the posts are initially acknowledged by the server. The temporary entry ID that ecto assigns is replaced by an actual entry number from MT. However I often get an error about not being able to post. Not always—sometimes they do go through properly—but not often. These posts show up on my website as well and they include the category. Usually if I try to post again, it goes through properly. It will tell me that I have the option of a new post or editing the existing post. I select Edit Post and it goes to the server correctly. The post is retrieved from the server and the pings are sent out.

I’m wondering if anyone else is having or has had any similar problems with ecto—even ecto 1.x. And I would really appreciate any suggestions or fixes that anyone encountered along the way. It seems like a really nice program but these continual errors and always having to check to see if the post actually made it to my website kind of defeats the purpose of using it.

Blacklist 2 Is Awesome

Thank you, Jay Allen, for creating such an amazing product. MT-Blacklist 2 is the coolest plug-in I’ve run on Movable Type—and I’m running Markdown which is super-cool in its own right, so I know how great plug-ins can be. In the 5 days since I started running it, I’ve blocked 231 comment and trackback spams. Prior to this I had moderation turned on for my comments so the spam never showed up but I’d still have to delete them from the database. And there was nothing I could do about trackback spam. It would show up on my site for as long as it took me to find it and delete it. With Blacklist I don’t have to do that. It just doesn’t show up. It’s truly great stuff. Plus with the auto-update feature I always have the latest list of ne’er-do-wells and blackguards.

It already has some pretty great hooks into the MT3 software. The entire interface is accessible from within MT for configuration. Plus little things like the link to submit checked comments for inclusion on the spam list make it a great addition. I can only imagine that it will be even better with MT3.1.

I’m A Fiddler Crab! Why Don’t You Shoot Me?! It’s Fiddler Crab Season!

In honor of the new Blacklist 2.0e, it’s open season on comments over here at MFTI. I’ve turned off moderation. I’m hoping that will encourage more people to join in the discussion. I’m keeping my fingers crossed that Blacklist will be able to keep out most of the spam. Of course by signing up for a TypeKey account you can post comments to this weblog and many others around the world, quickly and easily bypassing the spam protection. So it’s up to you.

Christmas In August


Toys, toys, and more toys. And version 2.0 toys at that!


Not only do I now have the new MT-Blacklist 2.0e installed on my server, but I’m also writing this entry with the new ecto 2.0b3 software. Adriaan was kind enough to choose me as one of the select 50 commandos to put his software through its paces. I will admit that right now it feels like I’m sitting atop a strange beast. I’m normally a “type it up in BBEdit and paste it into the web form of Movable Type” kind of blogger. I’ve never used ecto before. I really have no preconceived notions of what it’s supposed to be like. I have been interested in finding a way to keep a local copy of posts. I was considering whipping something up with Filemaker Pro and AppleScript. Something to make finding and linking to previous entries, since I do that a lot, was the idea. I think I can do that with this program, but I’m not sure yet.

So never having used this software all these windows and drawers seem a little unwieldy at first. But it’s also an exciting beast. For one thing it seems to already figured out that I have Markdown and Smarty Pants installed on my server. It’s giving me the option to choose them as formatting options. That’s very slick.


Ok, I’m already getting a little weirded out by the way these images are getting inserted into my post. Obviously I need to spend a lot more time playing. It will certainly be an interesting few weeks!

Update: 11pm —

I’m having problems posting with this. I’m still testing it.

Update: 11:45pm —

Another test with some new settings. Again. Again.

Update: 11:55pm —

It’s now working properly. So here’s the deal: I was trying to post using ecto but it was giving me this error:

Parsing failure!

Could not parse response for “metaWeblog.editPost”. Please check the console log for more information.

The weird thing is that the posts were showing up on my website. It seems to have been related to the fact that I’m running my CGI scripts in a security wrapper. I followed the steps on a page that Adriaan directed me to and now things are working correctly.

Local Virtual Hosts For Development

This is brilliant. Setting up Apache with virtual hosts on a local machine that mirrors a live server is a great idea for development. I’ve always setup other directories on the server, not accessible unless you knew what URL to enter. But with this method you wouldn’t even have to be online all the time.

I could take my laptop and sit out in the park and work. Actually, now that I think about it…