Category: Linux Tips/News

Do not be fooled by the provocative titles, these were informed and well done. Bryan Lunduke has given talks on this subject at LinuxFest Northwest. What do you think of them?

• Linux Sucks 2013
• Linux Doesn’t Suck 2013
• Linux Sucks 2012
• Linux Doesn’t Suck 2012
• Linux Sucks 2011
• Linux Sucks 2010
• Linux Sucks 2009

Our next meeting is Sun, October 13th. We meet on the second and fourth Sundays of each month in Berkeley near the Downtown Berkeley BART station. We hope you join us at Bobby G’s Pizzeria.

More and more people are using computers for their own needs in daily life, especially mobile devices with touch interfaces like smart phones and tablets, yet it seems like a smaller and smaller percentage of these users seem to have interest in learning how to write their own software, let alone have devoted the time required to build up skill in software development. This tradeoff in trust is actually quite remarkable when you think about it, especially if the source code you rely on is unavailable.

There is also a move away from running software locally on one’s own hardware to using software via the Internet and the World Wide Web. Software like wordpress doesn’t run on your own computer, it is simply accessed through a web browser. The "instant on" internet devices touted so many years ago seem to have remade themselves and are making a comeback after several cycles of Moore’s Law type formulations (wikipedia) have made them more powerful and capable. A backpack of batteries or a gas powered generator is no longer required to power these mobile devices for a reasonable amount of time.

What do you think? Does the shift in emphasis away from locally installed software make Linux more or less relevant in today’s world?

We meet on the second and fourth Sundays of each month in Berkeley near the Downtown Berkeley BART station. We hope you join us at Bobby G’s Pizzeria.

This was a question I saw on the linux4noobs sub-reddit on Reddit.com and i thought the question and answer would be of interest to Berkeley LUG.

Q: I am reading about disk partitioning and realized I don’t understand something I thought was very simple. This guide I am reading goes into detail about how /boot and /home should be on “separate partitions”. I know when I’ve installed linux multiple times I’ve had to “partition” the hard disk with a primary and a swap, and the ex4 filesystem gets written within that primary partition. Doesn’t the entire file system, including /boot and /home, reside on the primary partition I’ve formatted as ext4? Are these really separate partitions, or are they just different folders under / (root)?
Everything I know is a lie

A: The directory structure is kind of an abstraction.
It is a tree that lives under “root” or “/” and branches from there as sub directories.
Except, not quite. Partitions are kinda pinned (Mounted) to a branch node (mountpoint) of this tree. Then the sub directories from there are read out of the filesystem of the partition.
It all has to start with something mounted at / But from there on different partitions can be mounted at different points in the root filesystem.
So an example:
I have a partition sdb1 This is the partition supposed to be /home.
So inside we have directories for users, /tom, /dick, and /harry And each have the usual subdirectories like /Desktop, /Documents, /Videos etc
So we have sda1 mounted at / and it has everything in it. It has /home and had a user sally, so we find the directory /home/sally in there.
Ok now we mount sdb1 on /home. What happens?
The file tree starts at / which is sda1 and follows around insde sda1 until you hit /home where we jump over to sdb1 because it is mounted there (it is pinned to that branch)
So we get /home/tom and /home/dick and /home/harry because /tom /dick /harry are in the root directory of sdb1
And /home/sally is not there because it is in sda1 and after we get to /home (the mountpoint of sdb1) the tree traverses sdb1 instead.
The example was to show what happens, but normally you set things all up in the beginning and you don’t have sally go missing.
And creating and deleting would happen in the partition that is mounted there. So if Sally comes up and “hey I need an account!!!” creating /home/sally would make the directory /sally in sdb1.
Why?
Well It is kind of nice having all files in the same structure instead of the C: D: X: Y: Z: as in windows.
The whole /boot /home on separate partitions thing. (I may get jumped on for this — some people are very passionate on this topic) The idea is to have the greatest likelyhood of a workable computer when things go wrong. Like you do something stupid and mess up, Then you still have something working enough you can get in and fix it. Or that your personal files are on another partition and so protected.
Nowadays, The Ubuntu way is common: one big partition mounted on /
And If things go wrong you just boot a linux live and fix there

We meet on the second and fourth Sundays of each month in Berkeley near the Downtown Berkeley BART station. We hope you join us at Bobby G’s Pizzeria.

A challenge with all commons arrangements is the care required to maintain the shared resources to avoid tragedy of one type or another. In the case of digital "Free Software" and open source software, while there are important needs in caring for good software. These are sometimes not well understood by the developers and even less well understood by the users.

The analogies to common land only partially apply because of the digital nature of software. Software is now distributed over the Internet which is a hybrid public/private, shared, world wide infrastructure. A copy of a software work makes it more widely available with no impact on the original. If copies are used widely and code improvements are given back to the maintainer/distributor of the software everyone benefits from improved software and there is no degradation. Physical goods like land degrade when shared but software benefits when shared. One of the best examples of an information common is wikipedia.org.

A challenge is that enlightened self interest over a short time period does not illuminate the entire ecosystem of software. This can result in small decisions and free riders that endanger the sustainability of the shared common. Stewardship and collective trusteeship can only be appreciated when viewing the system over a longer period of time. What do you think? Please comment below.

We hope you join us for our meetings in Berkeley at Bobby G’s Pizzeria. We meet on the second and fourth Sundays of each month.

I sense a trend. Cloud people and others now love "open" so it seems that open concepts are more popular right now than proprietary, closed source concepts. I think practice is still catching up with hyperbole. This has been called open washing. Tim O’Reilly has been warning for years about creating data silos using open software for services. It’s not just about code any more.

There are also troubling disagreements like the tangential renaming of the wikipedia article about open source software by non open source grammarians that insist on a hyphen when it is not used this way in industry despite it’s more proper grammatical approach.

If people can’t even agree on the terms then erosion occurs from the more pure ideals of free software and open source software when describing them to others. It dilutes the effectiveness of proponents of these ideas so that we must explain complicated history to describe these anomolies along with educating people about the conepts.

Even the most dedicated, well run and financially stable proponents of open data and open source find open source solutions lacking for their own desktop use in their operations. Some of them are either unable or unwilling to take the (necessarily significant) time and expense to address their own needs with open solutions because it is seen as tangential to their primary mission. Any IT department that prefers open solutions finds that, whether they like it or not, they are forced to support some users and their proprietary software tools because there are gaps not addressed or not as effective in getting their real world work done. If open tools are not as effective then other solutions must be used. Who can fault them for making rational decisions?

Yet if nobody addresses these edge cases everyone loses. Until someone chooses to scratch their own itch for their own good then everyone who tries to follow in their footsteps will face similar tough choices. What have been your experiences with this?

We hope you join us for our meeting in Berkeley at Bobby G’s Pizzeria. We meet on the second and fourth Sundays of each month.