Today, I successfully completed a minor hardware project. This deserves italics because (like always) I've been feeling lately like I can't quite work on any projects, if for no other reason than I have a come-again go-again tendency to just fall asleep right when I get home. The particular hardware project that has been completed isn't too terribly wild. I installed some more ram in tacgnol, one of my old ThinkPads, bringing it to a grand total of 640 megabytes of memory.
Even with this additional memory however, tacgnol stalls out a lot when running. I attribute this partly to Windows Vista, partly to low memory, partly to a disk with a growing number of bad sectors, and partly to the fact that it is a ten-year-old ultraportable computer. This doesn't mean a lot today, but a decade ago, an ultraportable computer sacrificed a lot of horsepower in order to be as thin/small as they were, and/or to achieve a certain level of battery life.
The other really old ThinkPad I have hanging around, keskiviikko, is far more powerful and handily runs Windows 7 and one or two of the Office 2010 applications, it would even run Visual Studio 2010 if I asked it kindly enough, but it has been developing some issues starting up after a certain amount of time off of wall power. The natural inclination is to move the disk and wireless hardware from keskiviikko into tacgnol, but there's a point at which I need to acknowledge that they're each a decade or more old, and neither are particularly practical today. Keskiviikko is absolutely a brick, and even with its now-maxed memory and a faster disk, I just don't think I could trust tacgnol to move quickly enough to justify using it for anything, especially given that it lacks keskiviikko's excellent speakers.jj
So what's a guy to do? Honestly, I think it's time to figure out when it's time to give up on some of these older machines. As it stands, the next oldest ThinkPad I have, vorlaeufer, has hardware problems of its own and even though it's far faster than keskiviikko, manages to be even less portable, just because I need to use it in conjunction with a cooling pad (which dominates one of its two USB ports, and is noisy to boot.)
The hard question is what to do with these machines. I like them a lot and the temptation to just keep them and repair or restore them later is pretty significant, but there's also the option of selling them to somebody who has more patience than I do, or just recycling them. One of the issues I suppose I have is that it's not just these three machines – it seems like almost my whole life is filled with machines that are almost completely functional except for this one little detail. Even superslab, which I bought new just four years ago, has some issues I need to address when I get it back from the housemate currently using it.
I have eisbrecher, the Sony Vaio which is as of yet less than a year old, but let's be honest: I don't like it as much as I like my ThinkPads. It was easy enough to "move on" from superslab when it started showing some hardware issues, but as I look upon the housemate who is using it, I often wish to swap them out, if for no other reason than I believe I can fix that machine's issues. Of course I'm not just going to toss it, but who knows what I'll be using it for in a few years. Eisbrecher is honestly very fast, very portable, and very capable. I would be absolutely remiss in failing to recognize these things about it.
Given that I have eisbrecher and that everybody else can get by with just a single mobile computer, I suppose the question is why I still have a veritable pile of broken machines and support parts for them. Even if I take care to use them just so, is it worth my time and effort to keep old machines that while capable of running modern applications, aren't capable of doing so with any kind of reasonable performance or consistent frequency? I can't write a novel on tacgnol, draw up some numbers on keskiviikko or create a large flowchart on vorlaeufer if I can't trust any of them to even be running the next day, especially if I disperse them in such a way that they can't run daily backups to TECT.
The worst part for me is maybe that this isn't just about decade-old machines. Vorlaeufer is "just" seven or eight years old at the moment, and superslab is only four years old and does literally everything I need on a desktop computer. At what point is it reasonable for me, as an individual, to give up on machines? Should I set a cut-off at a specific year and review it on a regular basis, or set it at a specific age and review machines as they reach that age? Do I set a single unilateral cutoff point and apply it to all computing (and possibly other) equipment I use, or do I set different age cutoff points for different types of machines and equipment? My printer is six or seven years old now, and one day TECT will be old, even if it's not broken and still meets my computing needs.
These questions are almost certainly more difficult for me to answer purely because I'm used to having an excess of machines. Even if I don't use them on a very regular basis, each of these machines has carved out its own specific use case, I use them in certain circumstances, even if there's a newer machine available. Because of that, it's hard for me to say "oh, well keskiviikko is dead but I need a machine to play iTunes, let's go buy a ThinkPad T430," because not just one but three different laptops can also perform those functions, along with MILVAX the mac mini and topham, the wayward Inspiron.
Certainly one task ahead of me is to consolidate my endpoint devices. With the recent reorganization of TECT (which I will write about in the future), my endpoints are going back to being actual computing devices rather than just thin clients, which makes it all the more important that they work properly and be relatively powerful. It may be that once I've identified the different tasks my desktop computers do and consolidated these machines, there will be a case for a new one to replace the stack of old ones, but at that point I really do need to get rid of said stack.