The GASP/Akismet Experiment

It’s not often I do an experiment based on a post I read on another blog but I decided it was time to give one a shot. In this case it was based on a post that our friend Ileane wrote (yes, she actually does sometimes write posts on here blog lol) titled 5 Popular WordPress Plugins You Need to Ditch Now! One of the plugins she talked about ditching was Akismet, which I’ve always kind of had a love affair with, and thus I had to confront her, nicely of course, about the recommendation.

Has the NSA spying gone too far?
greg lilly via Compfight

(Growmap Anti-Spybot Plugin) would get the whole job done without Akismet help. It was developed by Andy Bailey of CommentLuv fame who, interestingly enough, said in an interview I did with him in 2009 that most plugin developers shouldn’t start off by trying to go after Akismet, and years later that’s exactly what he did. 🙂 I wasn’t really sure about it, but I told her I was going to experiment and write about it; this is that post.

A brief bit of history for the uninitiated. There have been a lot of people that have complained that Akismet does two negative things. It can put people on a negative list and thus always have every post of theirs showing up in spam or even being deleted before it ever reaches the spam filter. I’ve always said I had never noticed it and thus it didn’t impact me, but then Gail Gardner of Growmap did an extensive test last year on it and found that some of these issues might be true.

I still dismissed it because Akismet has always done a premium job for me, so it seemed. But I was compelled to do this experiment, and here’s what I’ve kind of come up with.

First, this week I’ve had less spam showing up in my spam filter than ever before. That’s both a good and bad thing mentally because often I had legitimate comments showing up in the spam filter, and over the past week I’ve only had one show up. I don’t know if this means it’s deleting legitimate people who it thinks is a spambot or if this week most of the people that comment have gotten it right.

Second, once I started the experiment I checked the box to allow trackbacks because I wanted to see how it handled them. I did get a lot of those in two days showing up in the spam filter, but not a single legitimate trackback so I turned it back off quickly enough. No trackbacks since.

Third, let me mention the spam filter. I wasn’t sure what would happen if I turned off Akismet and spam came in, but bad messages will still go to the spam filter, so that’s a good thing.

Fourth, if you saw my post about my comments problem you’ll see that I emptied a large folder through PhpMyAdmin that was holding all these statistics from my Count Per Day plugin, which I’ve also inactivated. When I went back I noticed my second largest file was something called wp_commentmeta, and it turns out that’s the file of everything that Akismet collects on comments it’s passed through and denied. Supposedly the WordPress program is supposed to empty that sucker here and there, but mine had never emptied over the years. Since I’d inactivated the plugin I was also able to empty that folder, and now I have so much capacity I feel like I need to start writing more. lol Yes, you can empty that folder safely, even if you’re still using it. And it seems there’s no files being created or filled up by GASP (which I’m still trying to figure out how I got ranked #1 on Google for ‘GASP anti-spybot’).

In my opinion, the GASP plugin has provided some peace overall to this blog, and that’s not a bad thing. I think I’m going to keep things as they are until I see there’s a reason to activate Akismet again, which I’m kind of doubting. And I’m going to do it on my other blogs as well. See, I can learn something from others. lol

22 thoughts on “The GASP/Akismet Experiment”

  1. Hey, Mitch. I turned off Akismet about a year ago and did not look back.

    The only thing I’ve found is that with GASP out there, the proliferation of human spammers attacking my blog is growing larger and larger. Very annoying.

    I thought Ileane did a great job with that post, I also deactivated my Facebook comments because it just made sense 🙂

    1. John, I came kicking and screaming, and I don’t hate Akismet, but at this point maybe I don’t need it. I’m still wondering in some part of my mind, but one less plugin is still one less plugin.

    2. The reason GASP exists it that I was getting 1000+ spam comments per day and it was taking me almost two hours to go through them picking out the dozen or more legit comments from my best commentators.

      GASP blocked ~960 of every 1000 spam comments and NONE of the real comments. That is why I would never want to do without it and why it got so popular so fast.

      Most spam is spambot generated (automated). Frustrated spammers even came by and complained about “what we were doing to them”.

      John, your blog is getting more popular and spam increases. If you were using Akismet you wouldn’t be able to easily tell which were human and which were spambot; with GASP you can.

  2. Akismet has lost its most staunch supporter? YAY!

    Well, Mitch, the beauty of this post is that you validated two important things:

    1. You’re open-minded
    2. The ONLY way to find out if something works is to test.

    Kudos to you! Be sure to check out Simple Trackback Validation with Topsy Blocker



    1. Mitch, I haven’t lost my support for Akismet as much as decided that, for now, I don’t need it. That file accumulating all that noise, if you will, irritated the heck out of me, but it’s done its job all these years as far as I’m concerned. And yes, I will test stuff.

      Course, now you’ve just thrown a new plugin at me that I’m not sure what it does, so I’ll have to check and see.

  3. When I converted Success Building Blocks from a “static” (though WP driven) site to a more traditional blog, I installed GASP because I heard it was pretty good and had one other feature I liked over Askimet: FREE. Yes, I know, Askimet is too, but only up to a point. Now, I’m glad I have it and am considering using it on the other sites too.

    Thanks for sharing your experiment!

    1. No problem Grady, but I’m wondering at what point Akismet isn’t free, since it comes with WordPress.

    1. No problem Jhong. Overall I had no problems with it either, but that data thing did disturb me.

  4. I’ve read Ileane post a couple of days ago, actually I am not using any of this plugins excepts Akismet. Never tried GASP, may be I will, honestly Akismet is doing its job, about a week ago, I’ve developed 2 new blogs and accidentally I forget to configure Akismet API key on one. After 24 hours there were more than 250 comments. Regarding WordPress, I think Akismet is ok, but I am having more problems on other projects which are not based on WordPress, I saw something really strange that everytime a company or website is mentioned, it lock the post. However I those websites are using commercial API which pops up info-box that comment is suspicious. Returning back to the post regarding ways to speed up WordPress, ComLuv is one of the biggest resource killers, also Twitter and Facebook connect.

    1. Carl, I’ve never heard anyone say CommentLuv hogged resources. It has had problems here and there with its servers, but otherwise it seems to work fairly fast. And 250 spam comments; ouch! GASP might eliminate a lot of those; I have to admit that when I first loaded it that stuff almost totally disappeared.

      1. I found this accidentally, working on a dedicated server with only 25, but very busy website. So we decided to buy more powerful server. To make long story short, I recreated the whole server locally – operating system,cpanel, etc. I used tools that are metering ins and outs of resources.

  5. Hi Mitch! Well done. I’m glad you did the test and reported your results. I’m also happy that you figured out the PhpMyAdmin puzzle. I’ve never heard of Count Per Day so thanks for explaining.
    I’m testing the new CommentLuv premium now which has GASP integrated along with a cool Track Back Validation feature. So far that’s working out well too.
    It amazes me what lengths people will go to just to get links to their spammy websites. It’s really kinda sad…

    I notice some slowness with my blog and it might be caused by many factors (including my host and the fact that I don’t have a caching plugin installed) but I’m not going to blame it on CommentLuv. Facebook maybe 🙂 but not CommentLuv.

    1. Thanks Ileane; you put it out there and thus I felt compelled to give it a shot. Amazing the stuff one can learn when they experiment. However, I have to admit I won’t be using the new CommentLuv Premium; you’ll be seeing a post saying why in a little bit.

    2. Hi Ileane,

      For sure it is not Facebook, FB/Twitter connect can slow website a bit down as it is HTTP request. Just a guess, probably too many connections to database at the same time. Yeah, I just check it you don’t use compression and I think it will be a good idea if you add plugin – it will reduce 254KB. Adding Expires headers will also speed up website, some of the static content is set 5min, good practice is to use aggressive settings on static content, something like 1year in .htaccess.

    3. Thanks for the monetary contribution Ileane; that’s greatly appreciated and I’m honored! 😀

  6. Nope, I’m leaving Akismet as it is because it is still collecting a lot of human spam and that is something that GASP is not designed for. I accidentally turned it off on one of my blogs and in just a day I had to delete a dozen comments and that was on a slow blog.

    On wassupblog kismet can get over 20 comments a day and that saves me the hassle of having to deal with it.

    1. I knew you’d be coming soon Sire. lol Actually, I haven’t had that issue in the time I’ve been running the test. As I said, I’m not seeing anything, which has made me wonder if some legit comments might be being dismissed. But people know how to reach me easily enough so I figure it must be doing okay.

  7. Akismet was the only spam blocker that I used on my blogs. Of course, I never got more then 30 comments per week so I always thought it did a good job as I never seen good comment getting blocked.

Comments are closed.