The original FreeBSD Mastery: ZFS books recommended managing disks by labels based on serial number.
pool: vm state: ONLINE config: NAME STATE READ WRITE CKSUM vm ONLINE 0 0 0 gpt/WD-WCC4N0JSJDKF ONLINE 0 0 0 gpt/WD-WMC4N0M8NRXM ONLINE 0 0 0 errors: No known data errors
I do this on hosts with multiple storage devices. It’s great for reality, with copy-and-paste terminals. It’s terrible for an educational book. The brain cannot absorb this easily. I see two ways around this. There’s the method used in the original books:
pool: compost state: ONLINE scan: none requested config: NAME STATE READ WRITE CKSUM compost ONLINE 0 0 0 gpt/zfs0 ONLINE 0 0 0 gpt/zfs1 ONLINE 0 0 0
Easy to understand. A terrible example. Readers of the first book did this, despite the copious warnings not to.
It was suggested that I could use truncated fake serial numbers from different manufacturers.
pool: vm state: ONLINE config: NAME STATE READ WRITE CKSUM vm ONLINE 0 0 0 gpt/WD-WCC4N0 ONLINE 0 0 0 gpt/SEA-4N0M8 ONLINE 0 0 0 gpt/TOSH-9262 ONLINE 0 0 0 gpt/NCC-1701A ONLINE 0 0 0 errors: No known data errors
This is obviously fake. It also obviously represents serial numbers.
So, the question for my nonfiction readers is: how does the latter example stick in your brain? Is it not only readable but absorbable?
I strongly support using the latter example. It strikes a nice balance between being absorbable and maintains a point I believe was highlighted in the original book: it still allows the user to uniquely identify the disk(s) and their location(s) should something fail.
I name things which I can’t think of a better name for, “Bob,” which works fine for one thing, but rapidly falls into chaos when you need a dozen names.
Probably best not to be like me. For a LOT of reasons.
“It’s great for reality, with copy-and-paste terminals. It’s terrible for an educational book. The brain cannot absorb this easily.”
I don’t get the problem.
I also find the first option, the, aestethically most pleasing one. And on top of things, the difference between option 3 and option 1, I would agrue, is not that big. If someone has problems understanding example 1, they will not get example 3 either. It basically does the same, just, imo, less clear.
No, I really don’t get it!
I have a very different problem. You drop lots of wisdom here and there. And, ideally, I would read your books cover to cover. And often times, I really do read a lot of them. And I take notes, on a paper, noting the page number, and, the nugged of wisdom dropped in a paragraph.
That way I can find them, since my brain remembers roughly what was said here and there. Sadly, that memory is not perfect.
Now then, I often find myself implementing things. Great!
Fast forward a few months, I want to change something / redo things / implement it somewhere else. And again, I remember a lot of things, but I am sure, it is not all. Still, most of time I am in a felt or real hurry. So I grab the book, hope that i find something in my notes, go over what I see in the chapters (!), and maybe read a few paragraphs. Only in worst case I start rereading everything.
I know, I am a bad person! In a perfect world, I would be perfect, and, execute things perfectly. That is perfect memory, perfect patience, perfect planning, … – so, even in the face of doom I would not falter, I would follow reason the most stoic way possible. An embodiment of mind over matter. Or mind over animal. ( Or mind over feelings )
So, long story short, the exmaples r important (!).
The offer fast recap. Orientation “in”, where am I in the chapter. What was, what is.
And then – think of the UNIX koans; in a way – I would go forward, and, do the best I can do, imitating the master, with the limited knowledge I have at hand.
Now again, in a perfect, I would not have let my little knowledge of pointer arithmetics I once fought for, gone to waste. And, i could simply dig into code, and, whatnot.
I can’t, maybe one day I will, … – if time permits.
For now, your books, are, kind of a “cheat code”. Or a quicker way, todo things.
And a road, I pay for, in money.
And they are worth it!
So again, … imo: the first example is the best. It shows me what I have todo, when my train of thoughts – or the flow I am in – does not permit me to read everything again. And I thus fall back to heuristics. ( Being an animal that I am )
If the examples deviate from what u have written, I guess, there will always be troubles.
Well, that’s my 50c.
The third example looks the most clear to me. Reminds me that I should use serial numbers, but also makes it easy to distinguish the different disks.
I see pulse-26c8’s point, but for me that’s slightly different. When I’m trying to remember what to name disks, I’d probably be leafing through the book from the front, trying to find the example where the names are introduced and than reading back a little to find the part explaining the logic.
When reading the book front to back or searching for other half-remembered examples, the shorter, more varied names would be a huge help. I tend to read the ebooks on my phone (I have no e-reader), so there’s very little visual context, but I think this is also true for other ways of reading digitally. Being able to remember that, for instance, the Seagate disk is dying in this example while the WD was recently added, makes it a lot easier to follow examples for me than if it were drives WD-WCC[…] and WD-WMC[…].
Plus, I think it might be easier to refer to the disks in text, which would also make for an easier-to-follow reading experience.