Visitor commenting: “Also, how can you talk about the absence of any suitable replacement for Tomboy without even mentioning Gnote?”

A very good point indeed! And one I made in my rebuttal to this claptrap, Here we go again – why Mono apologetics suck.

directhex response: “I’ve already talked about Gnote in a prior post – and feel pretty vindicated about my “it’s still too young for primetime” sentiment given http://bugs.debian.org/534969”

So I guess we are going to the bug reports? Then. Oh, what’s that? That bug was already fixed and closed … 4 days earlier?

Well, it is true that gnote does have 3 bugs on that tracker. 1 is a wishlist. All 3 are resolved, by the way.

I guess we could take a look at tomboy, as it has the directhex stamp of approval, and bugs are good reason to keep software out of a distribution.

Objectivity: a mono watchword.

Tomboy has 295 bugs. 7 critical, 4 major.

Now I know how unfair it is to point to a single bug or bug tracker and pretend like it makes the project unsuitable for public consumption. I know this is unfair. My question is, shouldn’t Mr. Shields know that too?

Mono apologists love this one! Basically it boils down to “software patents suck because your code might infringe anyone’s patent, so every project is just as risky as mono, so don’t sweat us over patents.” Now, like all really good propaganda, this has a some core of truth: it is indeed true that software patents suck.

I despise it because not only is it dishonest, but because how easily it is tossed around by mono apologists.

I guess it’s because the argument starts of with some truth – explaining how idiotic the software patent system is in the United States – that people just disengage their brain, nod along, and agree with the entire argument.

Of all software projects, not all are equally likely to violate patents.

Both of these statements are true:

1. Project A, a large open source networking project may infringe on one or more Microsoft patents
2. Project B, a small open source ping implementation may infringe on one or more Microsoft patents

Microsoft has an Open Specification Promise on a wide swath of technologies that they promise not to sue anyone over, so long as you don’t sue them first. Implementing parts of .NET that are not part of the standard may infringe on Microsoft patents, because Microsoft was careful to limit how much of .NET they gave to the standards body. Internal emails from Microsoft show a careful deliberation over how much of .NET to standardize, and how much to make sure was protect by patents, specifically mentioning Linux and preventing the commoditization of the .NET platform.

This is true in a more general sense as well. Any software project covering a common technology like ping or telnet is much less likely to run into patent issues for many reasons: length of time technology has existed, the number of similar projects having worked in the same area without patent issues, etc.

Of all possible patent holders, not all present the same risk in likelihood of relevant patents

Both of these statements are true:

1. Mono may infringe on one or more Microsoft patents
2. Mono may infringe on one or more Dunk the Clown, Inc. patents

It is more likely that mono infringes on Microsoft patents, considering Dunk the Clown, Inc. is not the that has over 150 patents referencing the .NET framework. It is unlikely that any of Dunk the Clown, Inc. patents apply to mono.

Of those patent holders that do hold relevant patents, not all present the same risk of enforcement

Both of these statements are true:

1. Mono may infringe on one or more Microsoft patents
2. Mono may infringe on one or more Red Hat patents

It is more likely that Microsoft will attempt to use its patents against the mono project, than it is that RedHat will. Sure, Red Hat might go crazy insane and declare some sort of war on mono. Also, I might get invited over to Scarlet Johansson’s place for drinks and a night of passionate lovemaking tonight. /brb

Of those patents that are enforced, not all are as likely to be valid

Both of the statements are true:

1. Mono may infringe some random “one-click-shopping” patent, and it turns out that patent is valid in a court of law.
2. Mono may infringe some Microsoft patent, and it turns out that patent is valid in a court of law.

There are some things you can do to defend yourself if someone goes after you for patent violation. You can show prior art, for example. However, .NET is unquestionably a Microsoft technology. It is not unlikely that Microsoft holds valid patents covering .NET technology. While Microsoft certainly holds many patents that are invalid for a host of reasons, it is foolish to think that they don’t have some valid patents, especially on a technology they created!

Of those patents enforced and upheld, not all are as easy to workaround

Both of these statements are true:

1. Mono could be found infringing a minor Microsoft patent, easily worked around.
2. Mono could be found infringing a major Microsoft patent, required substantial changes.

I don’t know if you could say which of the two is more likely, but it is true that we know Microsoft carefully considered what to keep out of the standard, and discussed preventing the commoditization of .NET. It is not unreasonable to assume that patents underlying important concepts that would require substantial changes protect those areas that Microsoft considers crucial to keep proprietary.

Summary

I would love to think this would put this argument to rest, but I know it won’t. The truth of this argument is that although all projects do face some risks, all projects do not face the same risks. Anyway, at least next time Team Mono whips this one out you’ll be ready.

You want examples? Sure ya do.

Jo Shields’ and the Ubuntu community are masters at this: One example we’ve seen before: Banshee is a perfect case where several lies and distortions are still being repeated, despite being corrected multiple times. And, the FUD aftershocks from the disgusting attacks on Gnote continue.

I know it seems I’m harping on Mr. Shields – but the fact of the matter is that his blog produces the majority of the pro-mono talking points that I encounter. So it’s simple just to go right to the source in these cases.

Those are exceptions, though, not the rule, right?

I see we have a skeptic on our hands. How about the claim that mono is faster than C++? In introducing the Mono.SIMD libraries, Miguel de Icaza provides a clear chart illustrating the speed gains over C++:

Look at that speed! Mono is awesome rocks!

It’s funny to note that all the language around this graph tries really hard to establish a tone of serious benchmarking: “I wanted to implement a real game workload”, “did a straight-forward port to Mono.Simd without optimizing anything”, it’s based on pre-existing code, asks for others to evaluate and post thier experience.

Well, buried down in the comments, someone did evaluate the bold-ass claim that Mono.Simd might be around 4x faster than a comparable C++ implementation:

The C++ program was compiled in Debug Mode. It’s not a fair comparison. Here is a benchmark on Windows XP with C++ program compiled with MinGW 4.3.2 and Visual Studio 2005.

Reality has a well-known C++ bias.

See those long-ass bars at the bottom of every test? That’s Mono.Simd. Of course, Mr. de Icaza quickly moved to update the source code and post a correction, since the results were totally invalid.

Hahahahaha, No. Sorry, I cracked myself up there. Actually, Mr. de Icaza took the show on the road to a Microsoft developer’s conference:

Gotta get that good word out!