Set up team to evaluate web-hosting providers for WP.org recommendations

[demetris]

The problem is simple to describe:

When setting up a CMS-based website, two of the most critical—if not the two most critical—decisions are:

• publishing platform
• web host

For the first you can pick at random among the best-known offerings and you won’t go much wrong. (Or you can pick the most popular one, and, in most cases, you will have made the best choice! :-D)

But the second is different — or so my little experience tells me: Fame and familiarity do not seem to be directly correlated with quality in the case of shared web-hosting, which is the best type of hosting for the majority of WP users.

And that’s unfortunate, because bad web hosting can ruin the experience of web publishing.

In addition, evaluting web hosts is not easy for a single individual, even for individuals who have the technical understanding, because it takes time and also money.

We, the people involved with WordPress development, are in a better position to offer some insight into the problem: We are many, we have collective knowledge, and also collective experience. So, here is my idea:

• Set up a team
• Set up a set of criteria
• Start evaluating services for a period of two or three months
• Publish the final recommendations on the WP.org page (along with the criteria used)

What do you think?

[nacin]

This should be proposed as part of 3.org (not sure if it was consciously proposed as part of that effort).

​http://wpdevel.wordpress.com/2010/06/24/hiatus/

[michaelh]

-1

Assuming ​http://wordpress.org/hosting/ makes valuable revenue to help keep wordpress.org running, why mess with that?

[demetris]

@michaelh:

Assuming that is true (that is, the recommendations page makes revenue for wp.org), nothing says that we have to lose that revenue if we modify, expand, etc. the recommendations:

The services we recommend now are not the only ones with affiliate programs. Maybe there are services that are better for our users and have equally good affiliate programs.

(I know for a fact that the two shared-hosting providers I personally recommend these days—and that are among those I would propose for evaluation if this task gets accepted— both have good affiliate programs.)

And, in any case, what services should WP.org recommend?

Those that are the best for WP users or those that make the most money for WP.org?

Using the latter as an ultimate criterion would be problematic, in my opinion.

[michaelh]

Replying to demetris:

And, in any case, what services should WP.org recommend?

Those that are the best for WP users or those that make the most money for WP.org?

If you don't make recommendations then that question doesn't need to be answered. If a host meets requirements that should be enough for wordpress.org.

A small example:
We recommend a host, that host gets overwhelmed from too many users, we take away that recommendation. So what do you say to those unhappy users who want to blame wordpress.org for something over which we have no control?

[demetris]

But the wp.org/hosting page does not simply list hosts that meet requirements. See what it says:

“We've dealt with more hosts than you can imagine; in our opinion, the hosts below represent some of the best and brightest of the hosting world.”

And retracting a recommendation is not something bad in itself—things change!— nor something that has not happened before in the wp.org/hosting page. In fact, only a couple of months a recommendation was removed for a host that had been on that page for years.

[michaelh]

Replying to demetris:

See what it says:

“We've dealt with more hosts than you can imagine; in our opinion, the hosts below represent some of the best and brightest of the hosting world.”

If you are being recompensed, you have to say things like that. I would suggest that if you believed that statement, you wouldn't have started this ticket ;)

[mikeschinkel]

Replying to demetris:

A huge +1 on this. And your suggestion is so serendipitously timely too!

After the ​conference I hosted last week to advocate for businesses to use WordPress (in addition to just the blogging loyalists) I was approached by one of our two main sponsors ​A Small Orange web hosting (a.k.a. ASO) asking how they could provide an offering that specifically targeted the needs of WordPress websites developers and owners. Evidently they believed in the message of our which was that WordPress is not only already huge for end user bloggers but will soon be absolutely huge for business users too. (note, ASO is local to Atlanta and I've been a customer of A Small Orange for 5+ years.)

So I have plans to meet with them in a few weeks to brainstorm how they can add lots of value by making offerings that are WordPress-specific. I'm sure I can come up with some ideas for them but it would be so much better if the community developed a set of best practices and suggested enhancements and I could simply communicate all the great best practices to them rather than just brainstorm and suggest a handful of ideas of my own.

So YES, I am all in for helping with this. Would creating a wp-hosting mailing list for web hosts and people interested in development best practices for web hosts make sense?

[mikeschinkel]

Related: ​http://markjaquith.wordpress.com/2010/05/14/web-hosts-should-adapt-to-wordpress/

[andrea_r]

I can help out especially in the area of networks/multisite. There are some hosts actively disabling the function from working, and users need to know this up front.

@michaelh Bluehost is one I see mentioned quite a bit in complaints, with people saying something like "this was recommended by wordpress!" and now something won't work for whatever reason.

[gazouteast]

what Andrea says, and add me to the list for evaluators please (I love grilling pre-sales and support people ;) )

Finally, something where I can make a definite and positive contribution back to WP.

This type of research and eval (with and without affiliate program attached) is something I've been building up as content for a new multi-site, which has been stalled by what Andrea believes is faulty server configuration for 3.0, even though the early WPMU dev work under 2.9.2 worked perfectly.

Moral seems to be that just because hosting is WPMU 2.9.2 friendly, does not mean it's 3.0 friendly - even if pretty permalinks work on the home blog of a multi-site, etc etc. (Still trying to resolve this and not 100% convinced it's the hosts or server - which is why I think I'd make a good ferreter for a project like this).

Gaz

[mrmist]

I think that I'm generally -1 on this idea.

WordPress.com offers a hosted platform, and other than that I think the .org side would be best staying away from the recommendation of services.

It's a difficult area to be in. There are sites on the Internet whose only business is to recommend other traders, and even they have difficulty in doing it fairly. It's too wide open to change [1], abuse [2] and subjectivity [3], plus as MH has pointed out, how is fallout dealt with when a recommended site proves to be unworthy?

All in all, I think it would represent a very large amount of work for the service to be done properly, and that effort might be better placed elsewhere (IE in the core product / site).

[1] The quality of hosts varies over time, and a sudden influx of subscribers can be a bad thing.
[2] There are hosts that I see recommended all the time that I wouldn't touch with a very long pole, how are such cases decided on?
[3] Swayed by their opinions, the people deciding on the recommendations can unfairly vote a host up or down, leave disparaging comments, leave fake feedback, etc..

[chriscct7]

This should be on meta trac, not the bug tracker for the core project. Closing as invalid.