Update 28/7/2010 – Theme Footer Links
The sales page for Thesis no longer mentions footer links though the license agreement isn’t available in public.
The sales page for Headway still requires attribution, subject to the license agreement – but the Headway Theme license agreement is available for public viewing.
You must also not change or remove the copyright information within any theme style.css file. You may however remove credit text from the footer of the themes if needed pursuant to the terms indicated below.
Full marks to Headway… choice is a wonderful thing.
I would really like to see the Thesis Terms of Service in public but maybe we will have to wait until they have them fully ironed out. Full marks as the intent certainly seems to be there.
Thesis is now GPL compliant with a split license to match the Headway Theme changes earlier in the week
Plus earlier in the week Headway Theme
I hope the guys realise that they can’t place limitations on how a user modifies the PHP, thus any limitation on removing footer links would make the theme no longer GPL compliant.
Also if they were using some kind of copy protection based around PHP to limit the number of installations, customers can request unencrypted source code.
Link Removal Update 2 – 23/7/2010
Will Anderson asked Chris Pearson directly about attribution links.
It seems Chris has no intention to change the sales terms
Chip Bennett queried him on it
Chris claims it hasn’t been discussed and they might remove that requirement
I honestly can’t believe it hasn’t been discussed internally – that one link is a “big rock” possibly responsible for 5-30% of revenue. That would depend on what percentage of sales are for a single license, and what percentage upgrade based on either wanting to remove that link, or to change the link to an affiliate link.
It also has a significant long-term branding effect, and has an affect on search/indexation including a knock on to partners of various types – in many ways it could be looked on as a bigger elephant in the room than the effect on any other proprietary code being shared/reused.
The Thesis – WordPress GPL discussions have been going on for 2 years… the situation of that credit link must have been part of that decision process…
Regarding terms of service, it seems DIYThemes have no intention of publishing their terms of service in public, before a purchase is made – I wonder what Paypal think of that. Would the terms of service make it seem (in Paypal’s weird way of determining these things) that the developer license is some kind of pyramid scheme?
I think it is time to be 100% truthful with my readers
- I take someone’s copyright work
- Often I take the copyright work of 100s of people
- I access the source code of the software and reverse engineer it
- I then create software modifications and packages that modify the operation of the original software, often in ways not intended or approved by the copyright owners.
- The code I write patches the original software – it cannot stand on its own
- Not only do I do this on my own account, but I am a dealer & dissemination point, actively encouraging others to crack and distribute software as well.
- In the past this has included the removal of protection
- It has also included adding in my own layers of protection
- I offer no support for what I create
- I offer no guarantees – in fact I often warn people never to use anything I create
- I am part of an organization – not formally, but our activites certainly could be looked on as organized crime
The laws I break have been tested in court.
However as far as I can legally tell:-
- The individual parts of the software I “crack” are Copyright by their respective authors
- The overall copyright for the collection of software is owned by… Free Software Foundation
- Just like James Bond has a license to kill, they provide me with a license to hack, crack & distribute modified versions of their software, in whole or in part
That license is the GPL
The following are just my personal notes & research about WordPress, GPL, Thesis & all the legal decisions that might have some relevance.
There is some personal opinion mixed in, that is me thinking out loud – I have deliberately not included any final opinions and kept all the following highly disjointed. This is not intended to be a flowing argument for or against one side of the controversy or the other.
The falacy of the moment in the WordPress vs Thesis GPL discussions, is that Open Source Software has never had it’s day in court over a software copyright issue. That there are no teeth to the WordPress claims that themes & plugins in most cases should be licensed as GPL.
So I spent hours digging into legal stuff to satify my own insatiable curiosity
I am possibly the worst person in the world to be trying to explain anything law related, so please treat all of the following as just my personal notes of no consequence other than they take up space on my server, and are posted in public.
Opinions here are absolutely personal, quite possibly subject to change based on the weather or shifts in the moon, or the amount of beer in the fridge.
I would also like to mention in my last post I tried very hard to stick to the moral issue of ignoring the interpretation of a license that was the desire of the license holder.
- I compared it to comment spam which isn’t illegal
- I compared it to email spam which in theory isn’t spam if you comply to the letter of the CAN-SPAM act
The Thesis / WordPress debate has been going on for over a year in earnest, in theory with discussions happening not just on the surface, but also behind closed doors, emails etc.
As an example just after Studiopress went GPL, but before the commercial themes page on WordPress.org was announced, there was this interesting tangle between Matt Mullenweg and Brian Clark (read all the comments – lots of them).
People Hiding Behind Anonymous Google Docs
I just love people posting stuff anonymously which they can edit at any time on Google docs. Maybe I should track down who tweeted it first.
However this document suggests that copyright comes before the GPL
It seems to think there could be a fair use argument
My personal opinion… bull$hit
Statements like “All this code is written by me” then we hear that the comment code ripped directly from WordPress was added by a 3rd party developer with no credit. (see references in second paragraph Mark Jaquith post)
Important Note: Whilst Mark works on WordPress code extensively, he does not work for Automattic
My personal interpretation of copyright & fair use is you really need to credit the source to have even a thin leg to stand on, but my personal opinion on this is probably not that important.
What is probably important is the way for the last 2 years the commercial message from Thesis has been that WordPress (and by that I mean the whole ecosystem that is GPL compliant) is somehow inferior in SEO capabilities.
The whole notion that the platform is Thesis, you just happen to have to use it with WordPress (currently) in my mind really screws up any notion of fair use as tries to undermine the marketing message
Also to justify the fair use argument it would have to be shown that Thesis is a stand alone application, but more on that later.
Summary judgement partially upheld that not including the open source software license along with code taken from an open source project was subject to DMCA. It was only partially upheld because at the time the judge couldn’t determine damages as part of the summary judgement as there was no commercial use by the plaintiff to base a summary judgement on.
1. The code in question was sufficiently original to be entitled to copyright protection. While not unique to F/OSS code, this was a legal issue on which Jacobsen had to prevail in order to assert claims under copyright law.
2. While the JMRI Project made its code available for free, there was “evidence in the record attributing a monetary value for the actual work performed by the contributors to the JMRI project,” thus laying the basis for monetary damages.
3. The removal of the copyright and authorship data contained in the pirated code was a violation of the Digital Millennium Copyright Act, thus providing a basis for suit for that action in violation of the JMRI license.
A lot of the discussion around the WordPress vs Thesis potentially originate at least in part from this case before it was appealed, overturned on appeal & subsequently settled. At that time is was being looked on that Copyright Law couldn’t be used because the terms of the license were not sufficiently limiting enough.
Erich M. Fabricius, Recent Development, Jacobsen v. Katzer: Failure of the Artistic License and Repercussions for Open Source, 9 N.C. J.L. & Tech. On. 65 (2008), http://cite.ncjolt.org/9NCJOLTOnlineEd65
I suggest you at least read Section B. Implications of Jacobsen on Open Source Licensing which explains why the “Artistic License” was thought to be too open, but also why the GPL which is written in a much more restrictive way is much more likely to be looked on favorably for copyright claims.
The Thesis camp have been pushing for a declaratory judgement for over a year
Here is another analysis of Jacobson vs Katzer worth reading for clarity (it is quite short).
Hacking & Cracking
Why did I start off with hacking & cracking in this post?
First of all it was the closest analogy I could find to the way WordPress links together with plugins & themes, and what changes when someone switches from a default theme such as 2010 or Kubrick, to an alternate theme.
In my possibly incorrect way of thinking, that is just a big ass software patch – there is no runtime, but it is replacing functions all the same. It is effectively patching the source before it is compiled, with the compilation happening in real time.
WordPress GPL couldn’t “infect” something like CaRP (the RSS display & processing system) even if Antone bundled a WP plugin directly with CaRP as all the plugin would do is provide an interface to the proprietary software.
It also provides a convenient lead into other topics such as legal decisions about mod-chips from the UK.
The UK appeals court judge wasn’t interested in the number of lines of code that were changed, but what that enabled.
I haven’t quite worked out in my head how this “transient copying” becomes precisely relevant, but remember these are my public notes for others to do with as they please.
Another part of the decision, and this was at the appeal court stage in the UK is that complex copyright issues should be handled by specialist judges in chamber, without a jury – I suppose that is the equivalent to the declaratory judgement the Thesis creators are looking for. (I am trying to be unbiased in this)
I come largely from a computer games background where mods of various kinds, from just changing a few variables in a script all the way to total conversions are prevalent.
There have been various equivalent licenses to the GPL created specifically for game mods, but ultimately I don’t know of any game mods that have been sold commercially without the express approval or involvement of the original publishers or developers.
Game mods don’t need to ship with the original code, it would be quite easy to package things so that they just install into the original game directory, or look for the required files where an original game is installed.
Whilst game developers and publishers are more than happy for you to create mods, they want exclusive right to exploit their intellectual property commercially. You would never get away with fair use, and in most cases game mods don’t even link together in the same way as WordPress does with themes and plugins.
One of the biggest problems in determining fair use especially in regards to the WordPress codebase is because it is so vast.
Say the 2010 theme which ships with WordPress
Lots of the work was done by Ian Stewart before he was even working for Automattic.
Some of the code almost certainly, even in small amounts might have come from his Thematic theme which has multiple contributors whether just ideas or code.
Thematic in turn came from Sandbox which was the first theme with really rich semantic markup, and some of that markup code probably ended up in core, or inspired it heavily.
Sandbox was also one of the first themes to extensively use custom css, and promote the idea of child themes.
I don’t want to be unjust, I know K2 was (and still is) doing lots of great things as well.
But the ultimate claim to fame for custom.css probably goes to Semiologic.
For over two years, the Semiologic theme was the only WordPress theme on the internet that allowed one to upgrade his site without worrying about losing his changes to his theme’s styles, through the use of a custom.css. Did we mention this theme was a precursor for a number of things?
There is no “black box” development going on – all these developers have been exposed to each other’s code – code of various plugins etc, especially for things like SEO, semantic markup, tagging, options pages, custom widgets etc.
To argue fair use, you can’t compare directly against the core code – you have to compare against at the very least the whole WordPress GPL code base including themes & plugins.
As an example a plugin author at some time came up with the idea of using the post title for the meta title tag.
Another might have added a way to define a title using custom fields, and yet another came up with the first interface to change this in the post interface.
Once a prima facie case of copyright infringement is presented, my understanding is the burden of proof would fall on the defendant.
The alleged derivative must “physically incorporate a portion of a copyrighted work… [or] supplant demand for a component of that work.”
In my wacky line of thinking, whist a proprietary theme might not supplant demand for WordPress core, it would supplant demand for other components of the WordPress project, such as the 2010 theme, or great WordPress SEO plugins like Headspace 2.
John Godley has been working on WordPress for years… I am sure some of his code is in core… after all he now works for Automattic, but I am sure there were contributions to core before that.
It would be almost impossible to unravel whether there is Headspace inspired code within core, or other plugins which then inspired core features in some way.
Thesis definately supplants demand for Headspace 2, and even directly suggests in the sales message that SEO plugins aren’t needed with Thesis (which is why many SEOs still use Yoast’s Meta Robots Plugin with Thesis)
Thesis certainly supplants demand for other themes that comply with the GPL wishes of the WordPress project as a whole.
Themes copy large chunks of code from each other, including original themes such as Kubrick and 2010 – 2010 no doubt has some code from Thematic/Sandbox which probably have some code seen in various plugins, including SEO plugins.
I also came across this..
Comparative advertising causes a stink in the Court of Appeal – much to the Judges’ regret
WordPress is certainly a brand with a registered trademark… the sales video on the Thesis website certainly disparages most WordPress themes, suggesting code bloat, invalid code, speed issues etc.
I would argue that any theme or plugin published under GPL is part of the WordPress Project, no matter whether you have to pay for access to that part or not.
Years ago it would have been so easy to add SEO functions into the WordPress core but a decision was made that they should be kept external to promote choice.
Any claims in a sales page directed purely at WordPress core is in my personal opinion significantly misleading.
There is also an invalid claim that valid clean code would infer some kind of huge ranking benefit.
Valid Code For SEO = Sales Bullshit
In general most WordPress themes are pretty solid on validation, because the designers actually care about validation.
Most of the time it is either users using 3rd party plugins or being themselves sloppy with markup that might cause validation errors & warnings.
A good example of plugins that really suck for valid markup is the Tweetmeme WordPress plugin – the Topsy one is much cleaner.
Some of those errors are things like Tweetmeme or errors caused by humans…
However there is this:-
Now I know this markup used on Copyblogger pretty well, as I highlighted it in a blog post almost 3 years ago (before Thesis). It is the method used on Copyblogger for the multi-line titles which is quite distinctive.
You will see that the markup previously used contained HTML entities for the line breaks. The current code does not.
This is what I have copied from my old post – it is quite possible the entities have been messed up a bit due to different code plugins, but it seems relatively ok, though the BR should really be stripped out.
<a href="http://www.copyblogger.com/writing-for-stumbleupon/" rel="bookmark" title="Permanent Link to Writing for StumbleUpon:<br /> High Impact Content â€œAbove the Scrollâ€ in Four Easy Steps">Writing for StumbleUpon:<br> High Impact Content â€œAbove the Scrollâ€ in Four Easy Steps</a>
This is what is used on a similar post currently on the Copyblogger home page
<div class="post-9508 post hentry category-1 category-creativity category-editing post_box top" id="post-9508"> <div class="headline_area"> <h2 class="entry-title"><a href="http://www.copyblogger.com/writing-perspective/" rel="bookmark" title="Permanent link to The Foolproof Cure for Weak Content: <br />4 Ways to Get Some Perspective">The Foolproof Cure for Weak Content: <br />4 Ways to Get Some Perspective</a></h2> <p class="headline_meta">by <span class="author vcard fn">Ali Hale</span></p> </div>
This is what the code looks like from a search page (I know Brian uses Lijit but that doesn’t stop someone adding a parameter to pull up a search result)
<div class="post-426 post hentry category-social-media post_box top" id="post-426"> <div class="headline_area"> <h2 class="entry-title"><a href="http://www.copyblogger.com/writing-for-stumbleupon/" rel="bookmark" title="Permanent link to Writing for StumbleUpon:<br /> High Impact Content “Above the Scroll” in Four Easy Steps">Writing for StumbleUpon:<br /> High Impact Content “Above the Scroll” in Four Easy Steps</a></h2> <p class="headline_meta">by <span class="author vcard fn">Muhammad Saleem</span></p> </div>
This is what the same code looks like from a related link
<li><a href="http://www.copyblogger.com/writing-for-stumbleupon/" rel="bookmark" title="Permanent Link: Writing for StumbleUpon:<br /> High Impact Content “Above the Scroll” in Four Easy Steps">Writing for StumbleUpon:<br /> High Impact Content “Above the Scroll” in Four Easy Steps</a></li>
Just to show I am not making this crap up I just added a break to one of my old blog posts on my home page to see how it is being handled by my theme, Thematic which is one of those free theme frameworks that have been downloaded 200K+ times (but I use SVN) – one of the ones being disparaged in the Thesis sales video.
<a href="http://andybeard.eu/2637/internet-marketing.html" title="Permalink to I Use Aggressive Hype & Obnoxious Tactics To Fool People" rel="bookmark">I Use Aggressive Hype <br />& Obnoxious Tactics To Fool People</a>
Strangely the BR doesn’t make it into the anchor title attribute on a well coded theme.
- It would have looked ugly
- It wouldn’t have validated
- If HTML was making it into the attribute I would question code security
It seems the poor output also appears when used in the sidebar… I really hope the input data is being sanitized.
Is Valid Code Important For Ranking?
Edward Lewis followed up reasons it might be with lots of detail.
Alan Bleiweiss followed up with more details on why he doesn’t do validation checks as a SEO.
The sad part of this is that Matt Cutts for some reason has allowed his photo to be published on the Thesis sales page, immediately under the video making all the claims about how the clean
valid code enhances SEO.
It does however seem Matt is considering switching back to “vanilla” WordPress, though he doesn’t mention whether he will use the 2010 Theme or maybe a free framework.
I sometimes get a little confused listening to Matt when talking about WordPress, as he might not be fully aware which panels within his “write” interface were added by his theme, Thesis, including options for custom meta descriptions and custom titles.
Those are features that can be added by one of 50+ WordPress plugins, which is why he no longer feels he needs the All In One SEO plugin or similar (go for Headspace 2 if you want something created by WordPress core developers).
I also wonder (just curious) whether Matt paid for a developer version of Thesis so that he could remove the credit link.
My primary issues with Thesis SEO features such as adding meta nofollow on archive pages were apparently fixed with Thesis 1.7.
Thesis officially has a 30 day refund policy – that is probably sufficient for normal situations.
However you can’t expect users to understand the nuances of copyright violations, SEO claims that might not be as valid as they appear at first especially in regards to valid code etc.
Then of course before you actually pay any money there is no indication that Thesis has some kind of proprietary license different to the WordPress project as a whole.
So when I see people attempt to get a refund based upon the current licensing issues, it makes me sad that such requests are being refused, no matter how long after initial purchase.
As stated on our website, refunds are available within 30 days of purchase. Therefore you are not eligible for a refund. I want to assure you that DIYthemes in no way deceived its buyers. When you purchased Thesis you purchased software. The GPL is a license not a law. Customers are not liable as you can see in this tweet by Mark Jaquith of WordPress: http://twitter.com/markjaquith/status/18808836688
If I was a customer of DIYThemes, and for whatever reason maybe due to all the recent GPL debate you couldn’t live with yourself to continue using the theme, and wanted a refund, this is how I would play it based on UK law. How it applies where you live is up to you.
Possibly the goods are not fit for purpose based upon whatever you might wish to state. That could be based upon what you read above that you think applies, but lots of the above could very well be based upon flawed logic – even the validation errors on the Copyblogger site could be due to beta code or some specific customization.
This does not necessarily entitle you to a refund outside the refund period that is stated, but that does not exclude a manufacturer from having to come up with a remedy within a suitable period.
I have no idea how they can come up with a remedy for complying with the GPL – code can always be fixed, though it might be hard to prove a benefit in SEO performance claimed on the sales message which by its very nature takes a lot longer to evaluate than the refund period allowed.
The Which magazine has a good guide for understanding the Sale of Goods Act in the UK.
You have the right to get a faulty item replaced or repaired, if you’re happy with this (or if it’s too late to reject it). You can ask the retailer to do either, but they can normally choose to do whatever would be cheapest.
Under the Sale of Goods Act, the retailer must either repair or replace the goods ‘within a reasonable time but without causing significant inconvenience’. If the seller doesn’t do this, you are entitled to claim either:
- reduction on the purchase price, or
- your money back, minus an amount for the usage you’ve had of the goods (called ‘recision’).
If the retailer refuses to repair the goods, you may have the right to arrange for someone else to repair it, and then claim compensation from the retailer for the cost of doing this.
You have six years to make a claim for faulty goods in England, Wales and Northern Ireland; in Scotland you have five years.
Now apparently if you read the article, within the first 6 months of purchase it is up to the person selling the goods to prove that your claim is wrong.
So if you happen to be in the UK, and you have purchased within the last 6 months, and wanted a refund of your Thesis purchase you could just quote the sale of goods act and ask for proof that your reason… copyright/license compliance/uncertainty for instance is unfounded – as this matter hasn’t been to court, there is no proof that could be provided that I am aware of.
In the US I have no idea whether the UCC (Universal Commercial Code) could potentially be applied to software, or whether you might be able to argue that there is an Implied Warranty of Merchantability that based on this article on the FTC site might last 4 years.
Generally, there is no specified duration for implied warranties under state laws. However, the state statutes of limitations for breach of either an express or an implied warranty are generally four years from date of purchase. This means that buyers have four years in which to discover and seek a remedy for problems that were present in the product at the time it was sold. It does not mean that the product must last for four years. It means only that the product must be of normal durability, considering its nature and price.
I certainly wouldn’t advocate contacting your local trading standards office, the BBB, FTC, Paypal, Ripoff Report, Consumer Report etc unless you felt you had significant cause and weren’t gaining satisfaction by other means.
Let me emphasise I am saying DO NOT DO THIS – The Thesis people are nice people, you are a nice person, I try to be a nice person – it is only a few bucks and Matt Mullenweg keeps tweeting thaat he will buy you a free replacement.
Personally I think that is over generous of Matt and he shouldn’t bare the financial brunt if customers are dissatisfied for whatever reason.
My personal opinion on warranties is that as a marketer it just isn’t worth fighting refund requests where there is even a slightly valid reason. You might choose to block perpetual refunders but ultimately they are a buyer, and if they don’t buy from you now, they could still buy in the future.
In the past I have fought extremely hard to ensure people were universally given refunds after I was personally refused a refund on a particular product I paid for. In that particular case it wasn’t the primary product owner but his new business partner on that venture and their support staff trying to “save the sale”.
Guess what? That is how Zappos & Amazon grow their business… ensuring customers are happy even when something they buy doesn’t work out.
Conversely not givng refunds can have a negative effect on good will – for me any and all good will evaporates when I hear of problems with refund decisions.
As an example a large portion of this post I already had stored in my bookmarks for my own reference – I would never have posted them.
But there is a tipping point… and for me that is refund policy.
Just to reiterate… these are my personal notes, subject to change and not in any way shape or form to be looked on as some kind of legal advice.
This blog post is like my personal wiki on GPL stuff, if you quote me on something written here, just like Wikipedia it might not be there the next day.
I personally made the decision 3 years ago, having paid someone 4 figures to develop a WordPress plugin without using any 3rd party code in a “black box” environment, to publish it as GPL… because it was the right thing to do. There was absolutely no pressure to do it. I actually gave the plugin away for free as well, but that was a commercial decision.
I also didn’t abuse credit links in any way.
Most of the current crop of commercial theme developers were still giving away themes so they could sell links – some people are now selling themes with licenses preventing you to remove design credits… possibly to sell links or it may be just to add more leverage for an upsell to a developer version.
If you don’t pay for the upsell to a dev version, financial logic is that you have a discounted version of the theme because of the link, thus you are paying someone to also have a paid link on your blog to them.
Google hasn’t yet interpretted things that such a link is paid, at least as far as PageRank penalties go, but they could. From my personal perspective which is worth nothing it seems to be abused.
I refuse to respond to tweets as Twitter still haven’t removed robots.txt preventing me finding my own historical tweets
Thus I am going to respond here on the blog
Harassment & fraud? I don’t think so
I compiled a lot of research which hadn’t been mentioned anywhere else to my knowledge in connection with this situation, and as I am in Europe there is a little more European perspective. Lots of factors regarding copyright are international.
I would have also drawn on some German decisions as well but they were a little too hardware focused, and I wanted to get away from Linux driver/hardware related stuff as it just adds fud.
What the product claims to do, the way that is worded in relationship to the product it is meant to be enhancing/replacing and actually achieves has a lot to do with fair use, especially considering my own opinion that the WordPress Project is more than just the core.
I am someone who spent a lot of time up until Thesis 1.7 was released, when asked to do a site review, informing people that they should switch off certain Thesis features because they had the potential of having a negative effect on indexation.
I very clearly stated above that that was fixed in 1.7 (as far as I am aware, though the default options might not be optimal)
If the comment was in regard to the information on refunds… I don’t think I encouraged casual refunds just for the sake of it. I provided factual information to the best of my ability which I think is far better than the Thesis support referencing a Tweet.
They are the ones without a terms of service of sale before you purchase, and no visible license agreement.