What I was what kind of idiot puts their home into a sat nav, don’t these people know where they live?I live in East Sussex, if I’m in Edinburgh I head south and pick up signs for London/M25, and I know which of the A2, A20, A21, A22, A23 gets me nearest to home. If I’m in Cornwall I head east, and aim for the A272. If I’m coming from somewhere from which I’ve never come before I head in the general direction of home until I see names of towns I recognise as being near to where I live. Any fool should know the towns around one’s area, and where one lives in relation to them. Finding my way home is always easy.

If going from home to somewhere I’ve never been before, if I don’t have an idea of where it is in the UK (Scotland, Wales, North-East/West, West Country, etc) I’ll at least get myself a clue before setting off and then, as I’m closing my destination (mostly > 50% of the overall distance) the sat nav may come in handy. For example, I’ve never been to Northumberland, if I was required to get to, say, Alnwick, I’d head north and between here and the north side of the M25 I’d be juggling in my head the long distance between the end of the M1, on A roads instead of motorways with the long way round M6 and then crossing to Northumberland somewhere north of Wigan but before Carlisle. I doubt there’s that much in it (the way I drive) – fuel consumption or time but I’d probably go A1(M) as I’ve never been up the A1 north of York so the view would be different. If I had sat nav I’d not turn it on until, if on the A1(M), I got to Durham (Co. not city), and if going up the M6 I’d probably turn it on around the Manchester/Liverpool junctions, not knowing how I was going to cross the Dales/Penines. I mean, this stuff isn’t rocket science, our country is really quite small compared with many.

How do people who need the sat nave to get from their front door to Asda and back again every time manage to work out how to even operate the damn device?! This reliance on technology is causing brains to turn soup. The genetic branching of Morlock and Eloi from homo sapiens has begun, I believe it happened some time around 1950.

My new server hardware doesn’t have an optical drive, nor does it need it. It does, however, happily boot from USB.

I tried using Disk Utility under OS X to write the ISO to the memory stick, but for some reason I couldn’t mount the ISO, OS X said that it didn’t contain a valid file system – which was rubbish because I did get the ISO onto the memory stick and installing Debian with it was a piece of cake.

I also tried using the

dd

I found various suggestions around the web, but I couldn’t make them work. I eventually found UNetbootin, a project hosted by SourceForge. UNetbootin will extract the contents of an ISO and write them to a USB stick in exactly the same way that they would be written to a CD (which was the original intention with ISOs).

To make the USB bootable UNetbootin also adds some other code to the device, this through me when I booted from my USB memory stick. I expected to see a regular Debian boot screen but was instead shown something else, though it contained a menu that had all the regular Debian options. I chose install and it all went smoothly – which is how I’ve always found a Debian install.

So that’s a big thumbs up for UNetbootin, it’s available for OS X, Linux, and Windows, and it’s free. A big thank you to the authors.

This problem could be due to the fact that the hardware is quite new, and I’m having to run a custom Linux kernel; maybe the code for some of the hardware isn’t quite good enough for production use. Or it could be due to a hardware problem, I’ve already had to return one which died completely on me while partitioning the disk for a Linux install.

So I’m going to retire the Mini from Linux duties (if it still crashes under OS X then I can attribute that to hardware).

If the Mini is being retired then I need a replacement server, step forward a great deal from Hewlett Packard on a ProLiant MicroServer with a 250GB HD, 1GB RAM, Athlon II Neo N36L 64bit processor, £240 and get £100 cashback. It’s a good looking server, much bigger, however, than the Mac Mini, and probably less powerful, but it’s going to be a server so that’s not a problem, no 3d games here.

I’ve already added another 2GB of RAM and it has three empty drive bays behind a front opening door (no need to remove screws and open the case) for easy upgrade of the storage.

One thing it doesn’t have is an optical drive, though it will boot from a USB drive or memory stick.

Instead of typing <h1>A Title</h1>, you just need to type # A Title, for <h2>Subtitle</h2> it’s ## Subtitle. In addition, markdown doesn’t require you to remember all the HTML escape sequences for &<># etc., you can just type the literal character and markdown will take care of the code for the entity for you, nice.

For all the details about markdown take a look at the Daring Fireball web-site. See, there’s another great markdown feature, automatic link creation, rather than have me explain it, read what the documentation says about it.

Anyway, I’m hoping to be blog more often, and consequently I expect much of this stuff to become known, rather than having to refer to the web-site all the time. But, until this stuff is knowledge, here’s my quick reference guide.

The general stuff, switching modes, search and replace, those I pretty much have remembered. It’s the stupidly simple stuff, navigating around the document, that gets me. OK, so a modern keyboard has end/home keys and also page up and page down, as well as the cursor keys, but vi doesn’t know about these (though I’m sure it can be programmed with a macro), so to scroll large quantities of text can be very frustrating. I don’t use any of this stuff often enough for these things to be remembered through repetition, so I’ve created this page so that I can just refer here when I need to know!

My Mini is a late 2010 model, the Model Identifier is

Mac Mini 4,1

You can find the model identifier of your machine by selecting Apple menu -> About this mac -> More Info. The model identifier is shown in the hardware overview.

I don’t know which models this information will work with, but if you try it with a different model please let me know of your success, or any problems you had so that others may benefit.

Overview

My preference is for a net install of Debian. This way I have the most up-to-date security patches and fixes, and my machine should be good to go when I’m finished with the install.

I chose to keep OS X on my machine in a small partition in case I need it for a firmware update at a later date.

There are two major problems with putting Linux on a Mac – an Intel one at least. The first of these is that Macs have no BIOS, what they have instead is EFI. Linux needs a BIOS so that it knows how to interact with the hardware. We get around this with problem with a replacement EFI called rEFIt.

The second major problem is the hardware in these boxes is either new, and therefore not in the regular install kernels, or it is set up to work in some non-standard way.

These problems lead us to need a custom install disk for this hardware, it may be that, in time, install disks become available that overcome these problems. Fortunately someone has already done this work for us, this web-page has a link to custom net install .ISO that can be used to boot the system and perform the net install. It also has the kernel you will need to boot machine after install.

Step one

Use Boot Camp Assistant to partition your hard drive. You can find BCA in the Utilities folder within the Applications folder. You can drag the partitions and re-size to however you want, I partitioned mine so that OS X has about 20 gigabytes, which is plenty, and it leaves a round 300 gigabytes for Linux. This machine is now a mail/news/web/samba server on a home network, if you’re installing for a different purpose then you’ll likely need to resize the partitions differently.

I set the new partition type to FAT, it stands out better when it comes to partitioning within the Debian install. When Boot Camp Assistant asks for the Windows disk (why don’t Apple acknowledge that people might want to install Linux?!) quit the program, it’s already done it’s job.

Step two

Download and install rEFIt. This is another reason for keeping OS X on the system, the rEFIt files reside in the OS X partition, I haven’t investigated, but there may be other ways to have rEFIt on the sytem without retaining OS X, but this works for me.

Step three

Put the CD burned from the .ISO mentioned above (or linked directly here) into the drive and then re-boot. At the sound of the ‘bong’ (what is that sound called?) press and hold the ‘c’ key, this will cause the machine to boot from the CD. From the boot menu choose ‘Advanced options’ and then ‘Expert install’ it is important that you do not choose the regular install as this does not give you the option to install the custom kernel.

Step four

Go through the install steps as they are presented to you, take note of which /dev your ‘/’ partition is on – you’ll need this for when you install the grub or lilo. Any that are skipped are either not important, or you can deal with them post-install.

At some point you’ll be presented with a list of possible kernels that you can install, you must install the one that comes with the .ISO, it’s linux- image-2.6.35-stlee-mac-mini-i686, this is the one with the drivers for your hardware.

When it comes to installing the initrd/initramfs choose the generic option, the machine specific option did not work on my Mac Mini, I don’t know why, but this had me banging my head against the wall for a long time and I didn’t know why it wouldn’t work. I spent many hours trying to fix it, but to no avail. Just use the generic version, it works and doesn’t cause headaches or drowsiness.

I chose grub (it’s actually grub2) instead of lilo, on a previous install attempt I chose lilo and it didn’t work, whether lilo would have worked had I chosen the correct initrd/initramfs I don’t know. Grub doesn’t tell you where it’s going to install so you have, at that point, choose to manually set the mount point, and enter the device you noted earlier. I don’t know how this will work if you are using a separate /boot partition.

Step five

Before completing the install and re-booting you must edit the file /etc/defaults/grub and change the line GRUB_CMDLINE_LINUX to include reboot=p nouveau.modeset=0 after which you need to execute a shell, and chroot into the install environment and run update-grub. If you forget to do this then you can boot the machine in rescue mode with the install CD, and execute a shell in the install environment and then carry out the above (which is what I had to do). If you don’t do this then part way through boot you will lose your display! The machine will be up, and all services running, it’ll even be waiting for you to login, but you’ll have no display. The reboot argument enables the OS to actually reboot the machine, rather than just halt it waiting for you to manually power-down, it’s the nouveau.modeset that stops the display disappearing.

Step six

You can now eject the disk and reboot, but if you’ve changed the partition table at all (very likely) you’ll need to run/re-run rEFIt to get the new partition table read. To do this start Terminal.app and:

cd /efi/refit

sudo enable-always.sh

That should be all that is required, a reboot at this time should give you the rEFIt boot screen and an option to boot either OS X or Linux, if you choose Linux you should get a grub boot menu showing your Linux install as the default, a rescue version of your Linux install, OS X 32-bit and OS X 64-bit.

Done

That’s it. There were a couple of things that I didn’t pick up from the wiki.debian.org/MacMiniIntel page that led me to have problems with the install. One was that there is indication that you should use the expert install method (regular install has always been fine for me in the past). And the other was that I didn’t know to use the generic initrd/initramfs. If I’d known both of these I’d not have had any trouble at all.

The MacMiniIntel page is a wiki, so I suppose I could edit it, but it’s not written well enough for me to just append the necessary, and as it’s someone else’s project (having created the net install .ISO) I don’t want to tread on any toes. I’ve sent the maintainer an email but had no reply.

I hope this page is found without too much difficulty and helps someone avoid the pitfalls that I experienced. Please comments regarding your experience either way so that I may update the page.

My wife isn’t always available to sail with me, and friends sometimes have other commitments, so finding crew isn’t always easy. So I’ve been wanting to give single-handing a try for a while, but didn’t want to attempt it with another pair of hands just-in-case. So today I took my wife and she agreed to do nothing while I tried to single-hand.

There are some places where single-handing is easier, and some where it’s harder. The floating mooring we had at Fambridge would have been great, boat hangs to the tide, untie and off you go. Coming back, with minimal way on, up-tide, I don’t think it would have been too hard to pick up the mooring. But we’re in a marina, and as Kooshty doesn’t go backwards with much control (long keel), so we berth forwards. Getting off the pontoon was easy, the wind was just pushing us on so I untied and stepped aboard, prop-walk astern carried us out into the fairway facing the right way. My first real trial was tying up in the lock, but I managed that – though wouldn’t fancy doing it in a full lock. After that we had a lovely gently mosey around Pevensey Bay.

Coming back in I managed to get along-side in the lock again – it was going OK, then went all wrong but I managed to recover and got tied up without any embarrassment. Out of the lock and to our berth. I was taking it very easy, very slowly. We’re on the left of a finger, on the right of the fairway which means that when turning into the berth the boat has a tendency to drift off the finger pontoon. My intention was to nose in and then reverse, prop-walk and stern-way then carrying us down the fairway a little so that I could approach from the other direction so that when turning into the berth any sideways drift would be onto the pontoon not away from it. All was going well, nose in, astern, slowly back out and down the fairway. This is where it all went horribly wrong.

I pushed the throttle lever (single morse control) forward from neutral and waited for the prop to grip and stop our sternway… Nothing happened. I gave a bit more throttle, still nothing, uh-oh what’s this, we’re going faster backwards. Both gears I had were reverse! By now the bow was coming across the fairway and I was worried about swiping boats with my bow. I asked my wife to go forward to fend off. Luckily I didn’t hit anything, but by the time my wife got to the bow we were heading backwards, away from the boats I was worried about, and now into an open berth (fortunately between two finger pontoons with no boat on either). I called my wife back and asked her to try and fend off at the rear (I was still trying to steer and find a forward gear) but we had too much way on, we struck the pontoon, I have no idea how fast, it just felt too fast.

Thankfully we didn’t hit any other boat. The only damage I have been able to see on our boat is minor – luckily we hit square-on and our stern is straight and flat, the force was taken across the entire stern and not just a corner. I have yet to check inside the stern-lockers for anything not showing from the outside, but I will do next time I’m at the boat – I was a bit shaken up at the end of the incident and wasn’t able to think clearly enough to check at the time.

That our first time out this year ended so badly does depress me a bit. I’m not sure I’ve fully thought through the situation yet, and don’t know if I’d do anything different if it happened again. I suppose I need to look at the gear box and see whether, if the problem happens again, trying forward gear again and again is likely to cause it to ever engage, if not then it’s obvious, I just fend off until we stop – we’re unlikely to have much way on so hopefully we wouldn’t go far.

I can’t say that I hope you enjoy it because it’s not that sort of article. But it will make you think.

If a thirteen-year-old is able to achieve this, what a wonderful thing it would be to have done. Anyone who has the necessary skills to achieve this feat can only go on to great things. The self-reliance, the resourcefulness, the courage and the stamina. We should be encouraging these attributes wherever they are found, not wrapping those who show them in cotton-wool to protect them from themselves.

If this child is living in a world of fantasy then it should be obvious to anyone with some sailing experience whether the child has the necessary skills and ability. Mental health professionals will know by interviewing her whether she has the mental strength to attempt a challenge such as this. If she has neither the technical nor mental strength to attempt this then she obviously shouldn’t do it. But if she does have what it takes then let her do it. I’m sure it would be the making of her, as it has been for the few who have gone before.

The risks are a lot less these days than when Sir Robin was the first to do this. Satellites give near pinpoint accuracy of the boats or sailors location, it’s not impossible for her to be shadowed by another boat with support services aboard, who could pluck her to safety should that be needed. It isn’t as if she will be out of contact for months on end with no-one having any idea if she is even still alive. With communications as they are it’s quite possible for there to be minute by minute communication, even a video link. Experts could talk her through problems with boat systems, help with navigation, give weather reports. I admit that this is still a huge challenge, and it is certainly beyond my ability, but don’t deny someone the chance to do this based on some arbitrary age limit which doesn’t take into account a persons ability, mental strength, or maturity.

Let her go!

You might be wondering who this al3x is that his rules have any value, well he’s a lead developer at twitter, this years favourite fashionable software application. In addition to that he’s been around computing for quite a while, check-out the rest of his site for more details.

al3x’s rules can be found on the link above. I’ve condensed them below to a list of the ones I hold dear, and my reasons for holding to these rules. These work for me, and are likely to work for many people, if you’re just an average Joe User with a home computer that gives you grief more often than it should then perhaps you should consider some of these rules for yourself.

1 Use software that does one thing well. 2 Do not use software that does many things poorly. 3 Use a plain-text editor, and not a word-processor. 4 Use a password manager. 5 Pay for software that is worth paying for, but only after evaluating it for at least two weeks. 6 Do not buy a desktop computer unless you regularly do processor intensive work (video editing, audio editing, 3D rendering). 7 Use a Mac. 8 Use a Linux or BSD server. 9 The only peripheral you need is an external hard-disk for storage. 10 Store all data in an open format.

Software that does one thing well

This is very important. Most of our time in any program is spent doing pretty much the same sort of thing. In a text editor you’re editing text, doing so must be as easy as possible. A good find and replace feature is essential, it’s even better if it can handle regex (regular expressions). Similarly, an email program should be simple to use and not get in the way of sending and reading mail. It should have a great search feature to find that one message amongst the many, and also good filing options, and filtering to organise your mail logically.

Software that does many things poorly

It’s not so easy for me to expound on this, because over my twenty-plus years as a computer user I find I quickly ditch this kind of software. The old Netscape Navigator program was a bit like this, it was a web-browser, email client, html editor, news-reader and probably a few other things too. A program just cannot be all things to all (wo)men because we each want to use programs in our own way. It may have worked for some people, but most chose separate applications for each task.

Use a plain-text editor

So many people forget that it is the words that are important, what they look like or how they are presented is of secondary importance. The words are the message.

A plain-text editor does not distract you with choices of font, the size of the font or even the colour. There is no choice of paragraph alignment or indentation. What a plain-text editor gives you is a blank canvas on which you can put down your thoughts, whether they be a letter, an essay, your novel, or the code from a computer program. There are no distractions, nothing to inhibit the flow of your thoughts into the computer. And once you have input your thoughts then editing and refining is a clean and simple process too. Once that is done, then you might want to fire up Word (better still use LaTeX) to layout and format the document.

Use a password manager

Passwords, oh don’t you just love ‘em. I bet you’ve got one that you use everywhere, most people have, and that’s what I had. But it’s not good, in fact it’s really bad. If someone overseas what you type, or sniffs your password from an internet transaction they can access everything your email, your computer, possibly your bank account, anywhere you’ve used that password on-line – facebook, amazon, play, ebay, twitter, gmail, yahoo, your entire on-line life.

You know that you should have a different password for each login, but you don’t because it’s too hard to remember them all. That’s where a password manager comes in. You remember one password, it remembers all the others. On Apple OS X, for example, Keychain operates seamlessly with any well-written application and automatically fills in login details (user name and password) for web-sites, FTP programs, network logins, anything. You have a different password for each login, which is more secure (if one is compromised, it’s only one, not all), and you only have to remember the one password that gives you access to Keychain. Simple. Just make sure that the Keychain password is a secure one.

Pay for software that is worth paying for

This is a simple one. If the software is good, and you want to keep using it in the future, through different versions of your favourite operating system then it’s got to be maintained. It might have been written as a fun project originally, but it’s probably taken on a life of it’s own, and the developer has to divide their time between what pays the bills and this project… now, if this project actually paid the bills then it’d get more attention. So contribute to the developer so that they keep developing whatever it is and you can keep using it.

Do not buy a desktop computer

Unless you need all that CPU power (and most of us don’t unless we’re rendering 3D images, or editing video) get something portable. Then you can work wherever you want, wherever is comfortable, and on the number 93 bus on the way home when you suddenly have that brilliant idea.

Use a Mac

I’ve spent the last twenty years maintaining the software and hardware at work, that’s in addition to my management role. Since we ditched MicroSoft on the desktop I spend almost no time doing that at all! I can spend my time on more important tasks. If we need a new machine a MacMini arrives, I unpack it, plug it in and a user is up-and-running in minutes. I don’t have to spend hours chasing all over the web for up-to-date drivers, or finding the latest PDF reader and anti-virus-firewall-adblocker-anti-spyware gizmo.

Just get a Mac, you know it makes sense. <">In world without walls, who needs Windows?<">.

Use a Linux or BSD server

I know that setting up or running a server isn’t for everyone. But for those even slightly technically literate it is such a good thing to have running. While your laptop or desktop is switched off or asleep a server can continue to collect your mail, filter the junk, file mailing-list messages as appropriate, and send out automatic responses in appropriate cases. It can serve web-pages for your domain – OK, I don’t do this anymore, my up-link isn’t fast enough. Maintain a news-server, time-server, file-server. Host backups, a web-cache, automatically update remote web-sites, or take backups of them. In addition to those tasks mine is also an MP3 Jukebox which I access using Theremin from any machine in the house.

You only need a hard-disk

Don’t print, email! If you want to display photos then get them printed professionally or download them to a digital photo frame. You don’t need a scanner, you’ve got a digital camera. All you need is somewhere to back-up your data… and a Linux/BSD server will do a good job of that!

Store all data in an open format

If you want to keep your data it’s because you might want to refer to it again. Whether it’s photos, music, letters, dissertation, whatever. But when might you want to refer to it again? If it’s a few years time then can you be sure you’ll be able to read the file-format then? I know I can’t open AppleWorks documents very easily, or Claris Works, or SuperCalc, or WordStar. Who can be certain that in five years time you will have access to software that can open a PSD file?

If it’s important to you use an open format – an open format is one that no company has control over. For example, MicroSoft Word and Excel files are in a format that is not open, MicroSoft knows, and has copyright over, how that file is constructed. This means that you are reliant on MicroSoft to continue to support that format for your future access to that data. The same goes for PhotoShop PSD files, and many, many, other file formats. For a better explanation of an open format take a look at this Wikipedia page.

Formats that are open are not controlled by one company, and the methods of reading and encoding the files is open to the public. This means that anyone who wants to can write software to access that file format. The real meaning is that the format is unlikely to die.

There are open formats for most media types be it plain text, audio, graphic, or video. Here are some examples

Summary

When my main computer was a Windows machine I spent time making the computer work which ate into the time I spent actually working. Now that I’m using a Mac I spend all my time doing my job, there is no time lost to keeping the computer running.