Now I am the first to admit that blog commenting right now sucks. When a company like CoComment, which does a mediocre job at best, can earn a living just by tracking and seeking updates on the comments you post across the Web, there is a serious problem. Users have a lot of reasons to prefer centralized commenting tools as having two or three commenting accounts beats checking dozens of sites for updates.

But what about bloggers? What do we get out of the deal? Though I’ve kept Disqus here on IS for some time, on PT I’ve been bouncing around from comment solution to comment solution trying to figure out what is best for my blog. I’ve tried nearly every service out there and the only conclusion I’ve reached is that no one, repeat no one, really wants my comments.

That is, at least not bad enough to create a truly compelling service…

Good People, Bad Products

Before I begin the rant, I want to take a moment and say that I’ve had interactions with people at all of these services. Every time I have talked with someone at Intense Debate, Disqus and JS-Kit, it has been a wonderful experience. They have very helpful people on their teams that bend over backwards for their customers. It’s great customer service.

The problem is that, despite all of this wonderful service, the products inevitably fall flat. It’s like having a five-star waiter serve you McDonald’s. No matter how much you dress it up, no matter how enthusiastic your staff and hardcore fans are, it’s still just ground beef.

What are my gripes? Well, let’s take the services one at a time. Each has their own unique pattern of fail.

Poor Disqus. Of all the three I’m mentioning, they seem to be the best right now. Even though Automattic purchased Intense Debate, Disqus manages to both provide the best features and continuously out-innovate everyone else. The problem is that they are victims of their own success. They’ve been bit by the spam bug, hard.

Even though this blog doesn’t get anywhere near the amount of comments of PT or other sites I work for, it gets many times the comment spam and most it is traced back to Disqus’ inability to filter. Though they’ve promised to deal with these issues, there has been a spam problem for a very long time. It has just merely ramped up in the recent weeks.

The biggest gripe is that Disqus will often email you even when it has caught a spam comment, turning comment spam into email spam. Smooth.

ID has managed to keep the comment spam issue under control, but has managed to place head squarely up ass when it comes to customer wishes. Facebook Connect usability is by far the most requested feature on their Get Satisfaction forum, getting more than 2 times the number of comments than any other suggestion. So what do they do? They refuse to make it a priority and instead say that it is “currently not on our dev list.”

Never mind that Disqus, JS-Kit and even standalone WordPress (with plugins) have this ability. ID is now dead last in terms of features (if you value Facebook Connect) and seems content to stay there. Way to listen to your customers…

I could never really get JS-Kit to work right on my site so I can’t offer any good gripes. Importing the old comments required me to email the entire WordPress XML file to them, which in turn necessitated putting it on a hidden URL since it was too big to attach, and the commenting system, once set up, didn’t seem to work right.

Where ID and Disqus are both real “set and forget” it seems JS-Kit takes a lot of work to make it look and feel right. That’s not a bad thing, but the instructions aren’t clear and the speed bumps in getting set up made me wary about investing too much time in the service.

I installed it, set it up, found that the comments did not work anywhere near as nicely as I wanted (though on other JS-Kit blogs it does look better) and turned it off almost immediately. I appreciate the help that they gave me, but it was going to clearly be a long road and I wasn’t ready for it right then. Plus it seems they’ve had some reliability issues lately.

What I Want

I don’t think I’m a hard person to please, but I do have seven things I want from whoever hosts my comments. Some of these are “solved” problems, but others are features no one is offering:

1. Easy Setup: I should be able to install the plugin, provide my login information and have the system up and running with just some minor tweaking.
2. Attractive User Experience: The experience, from the commenter’s perspective, should be very attractive and almost seamless. Bonus points for making me like looking at the admin panel as well.
3. Good Spam Filtering: Give me good spam filtering, hold the CAPTCHAs. I won’t think less of you if you use Akismet. I’ll think you’re smart.
4. Import & Syncrhonization: Your service should import my existing comments and then synchronize everything with my WordPress database. If your company goes under tomorrow, I don’t want to miss a beat.
5. That Includes Replies: Remember, since 2.7 WordPress supports comment replies, I expect those to be imported and maintained. No reason my comments should become unthreaded when I switch systems.
6. Facebook/OpenID Integration: Let me use Facebook Connect to post a comment. This lets my comment form reach a whole new audience and helps both encourage feedback and promotion of the site. Likewise, show some love for OpenID and other login systems.
7. Video/Audio Comments: Granted, no one has actually left a video or audio comment on one of my sites, but it’s a nice feature to have and there’s a good chance we’ll be seeing more of this in the future.

If you can do those seven things, keep your service reasonably reliable/fast and provide some good customer service, I’ll probably jump ship.

I’m not that hard to please, I just want it all.

Bottom Line

Right now I’m sticking with Disqus. I’m sure my readers on PT are tired of the repeat changes and, since IS is already on Disqus it seems like a good place to keep it for now. Though I hope they can improve their spam filtering, at least it has most of the features I need and good reliability.

I don’t know if I’d call this an “endorsement” of Disqus, but it is at least the service I have the least to gripe about.

But be warned, once WordPress gets Facebook integration built in or the plugins for it become a little bit less flaky, I’ll likely be moving back to my own comment system. That is, unless Disqus (or someone else) can really WOW me before that happens…

I could barely contain myself. When I raced to download the application for my 21-day trial I practically had my credit card in my hand. I almost bought my copy outright but stopped myself to give the software a try. I’m very glad that I did.

Though Blogo certainly has a lot going for it, advanced bloggers are going to want to keep away. It’s feature set is just to limited for people that require advanced manipulation of text and even basic posts are more difficult to post in Blogo than they are in the WordPress or TypePad editing panel.

However, those are the problems that can be fixed, there are other problems that seem to be a bit deeper into the development culture of the product.

Brief Background

Traditionally I have been a MarsEdit user. The bare-bones interface gave me the tools that I needed to get by and not much more. Sure, it lacked the full visual editor, but its HTML was more than robust enough and the ability add and edit tags helped me work with my custom template.

However, in recent months I’ve been trying to use the actual WordPress admin more and more. MarsEdit is Mac-only and I actually have to blog from three different computers, a Mac at my office, a Windows one at home and a Linux laptop. I have to at least be able to function on all three.

Fortunately, after some tweaks to my CSS and my code, I was able to set it up so that I didn’t need Marsedit’s functionality but still enjoyed the fact it had more tags and tools than WordPress. Still, I’ve been working to get used to the default interface and have been using it exclusively on all my sites for the past few weeks.

But there is still a lot of reason a standalone blog editor appeals to me, at least for most days. Sadly though, Blogo isn’t the one that I am looking for.

The Good

On the upside of the coin, Blogo is a beautiful application that is fast, stable and capable. What it does it does very well. I was able to get it to connect to all of my WordPress blogs without any issue. The application looks very good, the interface is easy to understand and it fits well with the Mac, which is something MarsEdit doesn’t do as well.

Overall, the process of setting up blogs and posting is very easy and intuitive. I also really like the feature that lets me drag images in and resize them directly in the application. If it could combine that with a built in screenshot program, I’d likely be in heaven.

For the most part, you should have no trouble figuring out how to use the interface or what to do with the program. Furthermore, it definitely provides a very “Mac-like” editing experience. For the most part, it will fit neatly within any workflow you have going on your Mac, especially if you are a fan of drag and drop.

The Bad

The bad news is that the very minimalist UI leads to a very limited set of editing tools. In fact, Blog has fewer features than the default editor for WordPress. You can bold, italicize, underline, strikethrough, quote and bullet list using the formatting tools provided, but not much else. There is no way to insert the “More” tag, something I desperately need, (Correction: This function is under the “Format/Insert menu. I did not see it. It is pretty thoroughly buried, you can’t be in full screen edit mode to use it and is named “Page Break”. I simply did not see it.) no way to add headers (which WordPress does lack) and the image support while cool, is more limited than WP’s current image uploader.

Compare this to MarsEdit, which has dozens of additional options right out of the box and the ability to add/create more if needed. Not only is Blogo limited compared to Marsedit, but it is also inflexible.

But once you are done editing, the problems resume. The last step for me on one of my blogs is to add some custom tags to the post so I can have the correct images display on the home page. Currently, no blog editor has this function (I’m not even sure if it is possible) so I have to upload my posts as a draft and put the finishing touches on the post.

This is not a major deal for me, but Blogo has no means to set up a post to be uploaded as a draft (Correction: This was a mistake on my part, It is available under the “File” menu. Not where I expected it, but not so horrible I feel vindicated for not seeing it.) Marsedit, by contrast, has the ability to not just set that option per post, but to make it the default, a big boost for people like me that need to make 100% something isn’t published before its ready.

Finally, I did notice a few bugs as I used the application. For one, the preview function did not work as advertised. Though it did the temporary post and pulled down the template, there were two problems with the process.

First, the temporary post was shot out over my feed as it wasn’t taken down quick enough. This made me very nervous about trying it again. Second, the preview never would display the text of the post I was writing just the “The is a temporary post” text that it inserts in the test post.

Without a “MORE” tag, no custom tags, no “post as draft” feature I literally can not use the application for any of my blogging. I depend on all three of the functions and, without at least the last item, there is nothing that I can do.

The Ugly

The problem through it all is that I actually wanted to like Blogo. I have a strong affinity for quirky apps, which can be affirmed by looking at their icon (see above ), I love the idea of full-screen blog editing, especially for longer posts, and the application’s look and feel is great.

So, in an attempt to see if I was overlooking the features that I needed (or at least plant a bug in the developer’s ear for future work) I headed over the to the Blogo support forums to ask my question. I posted a simple question about uploading the posts as draft and, after two days of being up, no one has come to reply to it. The forums themselves are pretty dead right now, only about half a dozen topics, but other questions have seen answers.

Compare this kind of support to IntenseDebate, who responded to a problem I had within one hour. I realize that Blogo is the product of a small Mac development company, but product support, even to free members, is often the difference between an app that gets used or one that rots on the digital shelves.

Sadly, I think Blogo and its company, Brainjuice, have lost my business for good. I can wait patiently for the features I need to be added, I did with MarsEdit, but to not even acknowledge the question or the issue, as well as other more pressing problems, is not excusable.

Conclusions

In the end, Blogo is a flawed but loveable application that could really become something of value down the road. However, if its company were nearly as loveable as the application itself, I might consider waiting for that to happen, but as it is I’m going to rededicate myself to using the backend of WordPress and hoping for the best.

WordPress’ editor may not be the best in the world but it has the features I need and I can use it anywhere.

Besides, at least with WP I know who I can ask to get help when I need it…

Though most of the platform is great, the search has always been the pits. Unfortunately, even as media handling, plugin installation and other features have grown up and become big boy tools, WordPress search has remained as daft as a speak and spell.

However, Automattic has this nice habit of buying up companies that fix their own weaknesses. Need blogger profiles? Buy Gravatar. Commenting system sucks? Buy Intense Debate.

So, in that spirit, I’d like to suggest their next big purchase. Lijit.

It seems simple enough to me. WordPress search stinks, there’s no easy way to fix it, Lijit solved the problem for Automattic and is making a tidy sum doing it.

The pairing seems natural to me at least. But I could be wrong…

WordPress Search: Epic Fail

If you want to play around with this problem. Feel free to use this blog. Since this site is fairly new and doesn’t have enough posts yet, I haven’t changed out the default search engine.

Now, do a search for Hurricane. You’ll get a few posts that are relevant but this post should come up first (unless you’re reading this much later after I post it).

The reason is simple, WordPress search is stupid. It looks for the keyword in the articles and orders the articles not on relevance, but on date posted.

To a reader, this is completely useless. They are likely searching because they’ve seen the newest posts, didn’t find what they need and have to pluck something meaningful from the archives. This means that almost every search result starts off with the least useful articles possible.

Of course, I don’t fault Automattic or WordPress for this. Relevance is hard and though plugins exist to help improve the search function, they are limited in what they can do. The problem is that effective search requires a lot of server resources, more than what most WordPress hosts can provide.

With almost no middle ground between “fast and ineffective” and “slow but useful”, offloading the search function makes sense. Since it is rarely used and can really drain a server, letting someone else handle the problem is a great solution.

This is why many bloggers just use Google custom search on their blog, partner with the best, maybe get some ad revenue and leave the old search behind.

WordPress needs to fix this problem if they are going to become a one-stop solution, fortunately, Lijit has the answer.

Why Lijit Makes Sense

Lijit makes sense because, like IntenseDebate, it is already so closely tied with WordPress. Even though the focus of Lijit is supposed to be aggregating search across all of a blogger’s content, including the social sites they participate in, the primary use is still to search within the blog itself.

Though it can be installed on any blog platform, the most common use I’ve seen is on WordPress blogs and the most common function is to replace the WordPress search. It is that simple.

I can’t tell for certain, but I’d estimate that most of Lijit’s hits stem directly from how bad WordPress search is out of the box.

Rather than having a company make its own profit from your shortcomings, doesn’t it make sense to buy the company, or at least the technology, and take the revenue for yourself?

The only questions Automattic should have are A) Would it be cheaper to buy Lijit or to create their own technology or B) Would Lijit be a company that is small enough for acquisition, it having received many millions in venture capital funding.

Either way, Automattic has to get this search issue resolved. Though it’s been a back-burner problem for WordPress for some time, as blogs grow larger and WordPress is used more and more as a full CMS, the search rapidly becomes the weakest link.

Conclusions

The bottom line is this, Automattic has two options. They can buy Lijit or they can fix search themselves. They can’t, as a company, allow another company to reap profits from their own shortcomings. That is leaving money on the table.

In short, it is time to turn this search problem into an opportunity. I think this has the chance to be a real profit-center for Automattic and WordPress in general.

It is time to jump on board and try to fix this problem. Failure to do so could result in a solution that remains forever out of WordPress’ hands.