Keep in mind that the points in the second are the same as the points in the first, just with a different perspective.<p>I think this mostly works OK except for one area: package format. The disparate package management in Linux is indefensible because it weakens the entire development ecosystem. One thing that hugely hurts the package management in Linux is that package format disparity means that the tooling for package management is far inferior to what it could be. The Linux package managers could be the single greatest tool for developers, but instead we have increased fracture over the entire ecosystem.<p>One of the biggest problems is the inability for the package managers to integrate with language-specific packages. Gem/pip/cabal cause an extremely obnoxious fracture because they all have dependencies which can't be tracked without an OS package manager but at the same time the OS package manager can't integrate with them. If Linux could decide on a single package format and packaging tool then we could all start working on tooling to support integration of the different language packages. However, as long as there are still 3-4 major package formats and packaging tools that's not going to happen.<p>Added to that are the package managers reliance on specific OS constructs. Homebrew works but we really shouldn't have to use different package managers just because we're on a different operating system. Either the package supports that OS or it doesn't, but this shouldn't say anything about the <i>package manager</i>.<p>Aside from all of this is the difficulty in actually constructing the packages. Arch PKGBUILDs seem the closest I've used to an easy source->binary package but they're not mainstream. As far as I know debs still don't support natively bundling the source with the binaries and the RPM SPEC system is the single unfriendliest system I've encountered to quickly make a portable package. Moreover, the ties to shell script just continue to harm OS independent adoption.<p>I'm not sure what the answer is but I'm starting to think the only acceptable solution is a complete severing of Linux distributions with package management.
Kickstarter sounds like a viable option for bringing more software to Linux if only for the reason that it's attracting a lot of attention lately.<p>There is a danger of this making Linux a second-priority OS but it pretty much is already so I don't know if this is a huge issue. A team behind a proven software for Windows or OSX could say "Hey look we've made this great tool for Windows/OSX users and made a lot of money on it.. We'll bring it to Linux if we raise X amount of dollars". The Linux users are significant enough and generous enough (as proven by the humble bundles) that it would work financially and would make the Linux ecosystem explode with new variety.<p>Open source developers on the other hand might find it easier to raise funds and continue to hack away on their projects with a similar solution. The Ardour developer could go on an open-sourced version of Kickstarter and say "Hey guys I've developed this awesome audio software that a lot of people use. Here are the features I'd like to bring to the next version.... I need to raise 100K in order to be able to continue working on this."
I'm so glad that I watched the first of these, and I think I will also enjoy the second. Since it required a bit of searching due to the very short mention at the end, the "Vivaldi" KDE tablet can be found at <a href="http://makeplaylive.com/" rel="nofollow">http://makeplaylive.com/</a> , which cites a "target retail price of €200" for the tablet stack.<p>Related to this, I have been helping seed the torrents and distribution of a new flavor of Kubuntu called Kubuntu Active, which can be found at <a href="http://cdimage.ubuntu.com/kubuntu-active/releases/precise/release/" rel="nofollow">http://cdimage.ubuntu.com/kubuntu-active/releases/precise/re...</a> . Like the Vivaldi tablet mentioned, it's based on the KDE Active stack and is thus optimized for touchscreens. I'm going to try this live CD out on my convertible tablet PC, once I can find a CD to burn it to. I was already very surprised to find that six months ago my upgrades magically enabled multitouch, but at the time it seemed like the only multitouch gesture anyone had programmed for was pinch-to-zoom in KDE.<p>Since it might also be relevant, in the vein of "send money to support businesses which ship Linux," shortly after I bought my current Fujitsu laptop I found a nice company selling Linux laptops called System 76: <a href="https://www.system76.com/" rel="nofollow">https://www.system76.com/</a> . While they don't seem to be selling convertible touchscreens yet, they do make some very pretty and cost-effective laptop options. (Full disclosure: I am plugging them in part because I want them to stay in business, because I want to buy from them whenever I need my next laptop.)
The guy who gave this talk has a really fun weekly show that I've been addicted to as of late. The Linux Action Show, most recent one here: <a href="http://www.jupiterbroadcasting.com/18887/ubuntu-12-04-review-las-s21e06/" rel="nofollow">http://www.jupiterbroadcasting.com/18887/ubuntu-12-04-review...</a><p>They do a live show on Sunday and put the recording out the next day. Today's will be on in about 40 minutes from now (1:00P EST)
I came expecting the latter would be in response to the first by another person. Nope. It's the same guy taking the opposite perspective on every point in the first.