In contrast to the Purism machines the Gluglug has free firmware. As far as I know this is the only commercially available solution with free firmware.
It's also the laptop Richard Stallman uses.
I've been using a C720 Chromebook with custom SeaBios firmware and running Linux.
While not a beast, a Haswell i3 with 4GB isn't bad at the ~$300 price point. Throw in a 128GB ssd and it's good to go. It's been my daily driver as I'm unfortunately content with 'open enough' given the market.
My wifes old netbook is about to die. Unfortunately I only upgraded it with a 2.5" SSD less than a year ago. I was thinking about buying a chromebook for her but was hesitant, because I'd really like to reuse that SSD. Does the C720 come with standard 2.5" SATA SSD or is it mSATA/m.2?
I've been using my X200 since I got it. Not only has it survived multiple falls from a 1.8m standing desk, it's still perfectly serviceable for everything I've needed it for. Most Thinkpads released since then are markedly worse, plagued with low-resolution or glossy screens. They've also removed the LED indicator that tells you whether the AC adapter is plugged in and when the battery level is critical.
tl;dr: sure there are newer machines with better specs, but nothing better in ways that actually matter to me.
I don't know about the X200 specifically but older ThinkPads tend to be really good with that. I have an X220 and it is easily the most serviceable laptop I have ever owned. It was super easy to upgrade the memory and add an extra disk. The battery is replaceable without opening the machine up and so it's easy to swap batteries if you use your machine for very extended periods of time.
Yes, that's another reason I've avoided upgrading--I really appreciate the ability to swap the battery out in the middle of a long flight without a screwdriver.
Basically everything in the X200 is user-serviceable; you can even replace the keyboard without taking it in, provided you can find the parts you need. The only thing I'd be worried about is being able to find new batteries for it; worth looking into before you pull the trigger.
I would think that power vs battery life would be pretty high on the list of priorities for technically inclined users. I current-gen core i3 (like the 5010U) packs quite a bit of performance vs. the power it requires.
For me, first and foremost are the parts that I interact with... namely the screen, keyboard and mouse. TBH, I have yet to try a laptop keyboard I like... I use a model-m style keyboard from Unicomp (103-key US), and all laptops and most other keyboards feel like sponges. I've yet to use a touchpad that's nearly as good as a macbook pro's, and their keyboard is better than most. The rMBP screen is pretty nice too. That said, there's certainly a price tag attached.
As to the software, I could care less mostly. I survive in windows, linux and osx on a daily basis, if I have a command prompt (bash) and most of the same tooling available (thankful for git extensions for windows bringing all the gfw/git/bash/ssh tools with it), I'm pretty happy. I have also started using an natural scrolling everywhere.
All of that said, it's nice to see open options that start under $1k in price, even if slightly older hardware. I'm not sure how good/easy Trisquel is to use, I know it's based with dpkg, but without apt/aptitude on top of it, may be difficult for most people to use in practice (I couldn't find reference to this, just a cursory look).
What do you mean by ideology? You mean they care about ideals and ethics related to software? Is that a bad thing? Even to people who don't care about ideology, ideology matters because they are affected by the dominant ideology. Actually that's how ideology mostly tends to operate, through the quiet complicity of the people who don't care. So when you talk about software without talking about licensing, freedom to read and change source code, etc, that is how the ideology of proprietary software becomes normalized.
... and as a result, I end up with reliable, light, somewhat costly machines(1) that come from one or two vendors and provide a consistent if inflexible experience.
This is not the lose-lose that some people seem to think it is. One of the reasons closed-source few-vendor is a dominant ideology is that it actually satisfies the use-case for an awful lot of users. Meanwhile, in 2015, I still can't get a reliable basic audio(!) experience on my Linux distributions because the open-source community can't seem to decide on the right way to do something as fundamental as "pipe bytes to the digital-analog-conversion hardware."
(1) Costly in general. With the Gluglug running at around US $500, it's not winning the cost game.
If the default audio setups of several different distributions fail to work, that's unfortunate. In my case, Dell's preinstalled Ubuntu had working audio, and when I replaced that with a custom NixOS installation, all I had to do was enable PulseAudio with a oneliner. Is your audio hardware supported by the Linux kernel?
Hardware companies keep specs secret and sabotage the development of free software drivers. That's feasible because they only need to collaborate with Microsoft and Apple to support a big market share. If consumers don't care, then apparently this is what happens. But it's not just a question of the competence of free software developers.
My hardware is ostensibly supported, but I often have to killall pulseaudio or restart the machine to get audio playing on Flash in Chrome.
Part of my point, though, is it's not a question of the competence of free software developers.
It's a question of the capacity of free software developers in the long-standing market / legal ecosystem to deliver a reliable solution at all. The roadblock may be entirely structural and completely unfair; doesn't matter. As the end-user (speaking with the Queen's "end-user"), we don't care why it doesn't work; we care that if we buy this computer / OS it works, and if we install that OS on that computer it doesn't. So of course we buy this computer / OS; we have work to do, and "Make the ecosystem fair for hackers to implement a working and reliable pulseaudio" isn't that work.
Knuth, when faced with a world where he didn't have a typesetting solution to print his book, wrote TeX. Most people just put up with the crappy existing solution, because their goal is to write a book.
Well, Flash is a proprietary bundle, right? Who knows what it's doing? It may not be PulseAudio's fault.
Of course there's some causal reason why many are stuck using proprietary software. That doesn't invalidate the free software cause, and is not a reason to snark about how "they don't care about software, only ideology." That distinction is itself ideological.
Thanks... I looked through bits of their FAQ, and a few other places. The wikipedia article referenced only dpkg, not apt/aptitude. Though, pragmatically I load some not entirely free software on my machines (chrome proper, flash, and a few other things namely).
I still think it's cool to have open options... I've tended towards buying routers/phones that I can replace the software, even if I don't start off that route because I like having supported, updateable options past the production date.
I've been less concerned on the desktop/laptop front, but am actually getting more so.. given the lock-down methods that have been in the pipeline, I don't want to be stuck with the MFG OS... I've been looking into getting a couple C2D macbooks, so I can load linux on them to play around with... I'm relatively happy with my current rMBP, but the changes apple has made to the OS irk me more with each release.
Trisquel is as easy to use as Ubuntu. It's built from Ubuntu LTS with all the proprietary parts removed. The default desktop environment is a customized GNOME 3 meant to operate more like GNOME 2, but it supports everything Ubuntu supports, including standard tools like aptitude.
Just throwing this out there, but isn't it possible that a lot more performance can be squeezed out of a machine that offloads less to the CPU in binary blobs?
In the end it's all binary blobs that go to the CPU... in the case of closed software, it's that the source is unavailable, so you cannot confirm security, or patch bugs when you encounter them... tbh, I don't typically dig that deep into the bowels, but have had ever increasing concerns, that have manifested themselves more in the cell phone and home/small-office routers than anywhere so far, but with Windows RT came into the desktop world full front.
I want to be able to use my hardware beyond the manufacturer's expiration date. If I choose to replace functional hardware with new hardware, that's my choice.. but the ever encroaching state of planned obsolescence disgusts me. Which is why I really appreciate FOSS, and more open hardware options.
I'm typing this on an (unliberated) X60 which I use as a sort of netbook and which steadfastly refuses to go wrong! I'll keep it out of landfill as long as possible I think.
I'd argue that this is far from slightly outdated. Starting with RAM, four gigabytes of RAM may work for a barebones programmer machine, but once you want to start doing any sort of media work, like editing photos, video, etc, you start running into that wall real quickly. Even software development work is going to cause you to run into limitations. As the processor doesn't have an IOMMU, any vms you want to run will be limited to slower, virtualized, graphics, and that expresscard slot on the side becomes a security concern (hello, evil maid). Additionally, it only supports outputting to vga, limiting your options in monitors, and making sharing content on a friend's tv a lot harder.
So in short, its constraints in various areas really do shackle you in a lot of ways. Media creation, software creation, sharing, all suffer from the weaknesses of this machine.
Not everyone needs to do all of those things. In fact, I'd argue that very few people need to do all of them. So, yeah, it's a fraction of the machine the (not maxed out) 13" rMBP I'm typing at right now is. So? If "freedom" (as the term is used by the FSF and folks of similar ideologies) is your priority, you make some compromises. If not, some would argue, that's a compromise you're making, right there.
The x200 also unofficially supports 8GB of system RAM. Since the machine is under no warranty coverage whatsoever, running an unsupported configuration will, at worst, cost you the price of a couple of 4GB SO-DIMMs or return shipping on an RMA.
Please don't judge a thing that doesn't satisfy your priorities as being somehow objectively "bad" or "wrong". It just doesn't meet your needs. Go find a machine that does, then, and let the people whose needs and priorities are met by a Gluglug x200 satisfy theirs with this.
So I'm not allowed to discuss what these people are giving up in the name of "freedom"? That the fact that this is a refurbished machine from seven years ago shouldn't be brought up? That functionality was removed in order to, through some bizarre twist of logic, make the machine "more free"? Given that this is a machine that is built in the name of ideology, I would argue that questioning aspects of that ideology, and what is demanded to acheive their idea of purity, is more than fair game.
You're perfectly free to discuss it. Your posts read more like telling people they're wrong in finding these machines suitable for their purposes than just discussing the tradeoffs they're making. Even this comment, with clearly un-biased turns of phrase like "bizarre twist of logic" and "their idea of purity" conveys a conviction on your part that these people aren't merely wrong, but deluded.
Tone down your dismissal of other people's needs and use-cases, and we can have a discussion.
Remember, RMS uses a Libreboot-based x60, which is even less machine than the x200 under discussion here. Are you going to tell him he's wrong, too?
I feel that the machine here is borne more out of ideology than logic or necessity. The laptops in question, and the software installed, are designed to take away any ability to do things that the designers do not like. For example, the ability to update the CPU's microcode, or load firmware for numerous devices. I would argue the rational approach would be to not ship the firmware, but keep the drivers in place for a user to use them and the firmware, if they so desire.
Regarding RMS, I do not feel that he is a good spokesperson for the Free Software movement, or computing in general. He's requested people to delete Free software and pretend it doesn't exist because it could potentially be used with unwritten software to potentially allow a non-free program to use gcc as part of its chain [1]. He has criticized software because it allows interoperation with a perceived enemy [2]. Lower down on my list of criticisms would be his use of the x60, because it is in a lot of ways saying that free software is an inferior good. Instead of trying to work with a team to create a completely open laptop, like Bunnie Huang has done [3], or use a Chromebook that runs a Free software boot environment by design, he uses a very old laptop instead. Which in the end, is one of the biggest flaws with RMS's operations. He seems unwilling to show that one can use Free, modern, devices and still not limit himself.
I use sharp words because I am tired of the lack of vision, lack of foresight, and lack of creativity from the gnu branch of free/open software. Modifying a seven year old laptop and making it less functional is not revolutionary or interesting. Selling said item for a three time markup is not revolutionary or interesting. There are more than enough options out there, the Free software community needs to put up or shut up.
"but once you want to start doing any sort of media work, like editing photos, video, etc"
Its interesting how slowly use cases change and how slowly software changes, but how fast hardware increases. In the 90s I would have really been lucky to have 4 gigs to do those tasks you list.
For me, the larger fallback would be the largest SSD options seems to be 120GB, if this is a SATA2/3 interface, which I would presume it would be nice to see a 250/500GB option for SSDs which are generally more of a boost for me than going above 8GB ram.
It's the Intel "Calistoga" platform, which is very similar to the X60 in that it uses the same chipset (945) and Apple probably followed the Intel reference design closely since it was their entry into the x86 world.
IBM/Lenovo tends to follow the reference designs more than the other companies, which helps with Coreboot porting.
I wish more common laptops would have replaceable firmware, like Lenovo and the like. Having all those blobs with who knows what is very annoying. Especially when that firmware starts behaving as if it owns your computer, and refuses to boot because "WiFi card is not authorized" or some other such idiocy.
Yes! I bought a Lenovo laptop in 2008 and tried to replace the wifi card with an Atheros-based one (better driver support on Linux at the time), and it refused to boot afterwards. Never buying a Lenovo laptop again.
I got one recently, and it came with 802.11n Intel card. I bought 802.11ac Intel one (it has a pretty good support on Linux). And who knew that some FRU part code wasn't from the "authorized" whitelist and that nasty firmware would refuse to boot with it.
I had to return the card, and figure out what the compatible FRU part number is. The card is really exactly the same, just some internal code is different. This is beyond dumb and annoying.
Yea, I don't like Intel Boot Guard for that reason, and I think this is more important than being "100% FOSS" or the like. I once suggested allowing it to be disabled using a jumper or the like.
Physical write jumpers would solve 99% of firmware security issues and significantly reduce the attack surface of modern computers. APTs can't really manifest unless they come pre-installed.
Then again, NSA and other state actors wouldn't be able to implant malware into those areas without physical access.
Firmware these days requires more non-volatile memory in early init than is available outside flash.
So with a write protect switch, RAM init couldn't store the actual memory configuration (~2kb of data), so wakeup from suspend to RAM doesn't work.
In case of UEFI, its variables are also stored in flash, so that also doesn't work. Since everything optimizes for UEFI (as far as firmware matters are concerned), that jumper would be counter-productive.
I see here a _big_ security argument for libreboot. With free firmware and free kernel you can arrange suspend to RAM with write protected firmware memory.
A lot of systems use fairly off-the-shelf bioses, that have lots of tools available for them that were intended for OEMs to do the minor, annoying tweaks that you described. There are sites [1] that specialize in discussing and using these sorts of tools. I had a lenovo laptop that behaved a lot like that, and was going to use a modded bios to fix it, but discovered that an updated bios actually had a bug in the halt code, so it would claim to halt, but five seconds later would continue the boot process.
> Very neat, but it's specs are a little lack-lustre
It's kind of the nature of the beast. It's one of the only (or is it the only?) laptops they can do this with, so if you're committed to a FSF-approved machine you're stuck with old hardware.
> It's one of the only (or is it the only?) laptops
One of the only; there are several other models that are supported by Libreboot[0]. Of those, Gluglug used to sell X60, X60s, and (briefly) T60 Thinkpads as well.
Why is that? From the site I see that it's just a modified X200, wouldn't they for example be able to modify a more modern model? Does Intel have special measures to prevent that? Or is it simply reverse engineering that hasn't been achieved yet?
AIUI, it's primarily a matter of the machines for which libreboot support exists. So, for example, the X60 and its tablet version are supported, but the X61(t) (of which I have two in a box somewhere) aren't. This is a terrible bummer for me, because that's still one of my favorite machines I've ever owned.
Intel started embedding a processor in the chipset called the Management Engine. In vPro systems it powers the AMT features (remote control, McAfee virus scanner outside the OS control, various other things), but it's around everywhere.
The X200 contains the last chipset where this processor was optional - if it comes with no firmware, it simply disables itself.
Newer chipsets sometimes feature a recovery/service mode (up to 30 minutes of activity before the device shuts down, with limited capabilities, eg. no power management).
Now, this chip wouldn't be so bad, but unfortunately it only runs on firmware signed by Intel. Lack of signature means that the system doesn't run, or only in that service mode.
Since that project hasn't managed to cracked RSA2048 keys (and who did?), that firmware precludes using newer Intel chipsets.
It's been a long time since I've built a computer myself, but back in the day the Intel chipsets weren't the only ones that could boot Intel processors unless I'm mistaken there were always 1 or 2 alternatives. Or is this not the case for laptop systems? And what about AMD, do they do the same thing?
Intel CPUs are tied to specific Intel chipsets these days, with no third party vendors.
AMD used to be much better (they have some small firmware for a support function that was signed, but even that was so half-hearted that one coreboot developer managed to recover the key - but didn't publish it).
These days they support ARM Trust Zone (with an ARM processor), and that part of the system is all signed (as "Trust" doesn't mean that _you_ can trust it).
It's Hollywood, essentially - can't have you crack your movie-playing device to copy their precious bits. And what else would anybody use a computer for, if not for watching blockbuster movies?
That hasn't been the case for a long time (a decade?). Nowadays you can only use Intel chipsets, even in desktop systems. I believe it's the same for AMD.
modern models have the firmware encrypted + signed, so its a great deal harder to modify it, added in with the fact that intel requires several binary blobs to even boot modern processors...
(Look at the single-thread rating and you'll see that a single Core 2 core is 2-3x faster than a single Atom core.)
Unless you're doing something particularly CPU-intensive, which doesn't tend to be the use-case for a 12.1" ultralight laptop, that processor is more than enough. The small screen resolution may get a bit annoying though.
You're comparing apples to oranges here. Yes, everything you listed is an Intel x86 part, but the Z3735F is the CPU in my Windows 8 tablet, which has a 4W "SDP" (scenario design power).
The P8400 has a TDP of 25W, or about 6 times the heat dissipation of the Z series atom. So of course the Core 2 can beat an Atom, it's in a completely different category of processor.
If you are going to compare apples to apples, pick the i5 2410M. It's not a new part, but the TDP is comparable to the P8400, and the target market (12.5-15" laptops) is the same.
Here you can see that the newer laptop-class CPUs out perform the Core 2 by a factor of 2. Couple this with much better integrated graphics, and newer memory (4GB DDR2 SODIMMs exist but compatibility is spotty and they're quite expensive) and it's not really a contest anymore, the newer CPU/chipset provide better performance and will give you longer battery life.
I am not going to claim I am one of those people who needs a powerful processor in my laptop, because 90% of what I do is login to remote servers who do the work much faster than my laptop ever could. But, I do enjoy getting 5-8 hours of battery life thanks to my newer processor sipping electrons when idle, which cannot be said for Core 2 era systems.
I know there are more powerful CPUs available; the point is whether that CPU is sufficient for most of the things you're likely to use this laptop for. Is your Win8 tablet too slow, or does it do what you want with good performance?
For what it's worth, I couldn't find any information on the battery in these things. I understand these are second hand, but is the old battery replaced? For that matter, what is the quality of second-hand X200's anyways?
I would wonder whether or not the keyboard is worn down, or if the screens are still in good condition. Has anyone bought one of these that can discuss this?
I confirm. I had a machine similar to that one up to 18 month ago. It was fast enough to develop web applications (Rails, JavaScript). The only problem was the RAM limit at 4 GB. I replaced it with a ZBook 15 with 16 GB RAM. It's faster, so doing the same kind of work becomes more convenient, but the main advantage is not to have to close some programs to start another one. That's particularly handy when working with virtual machines or multiple browsers.
I want something like this, but for a phone. The Freerunner was the closest thing to an open phone, but the call quality made it unusable (speaker hissing and other noises).
The Neo900 project is the successor to the Freerunner/OpenMoko, and might be worth checking out. They're still in development, but it has a few pretty compelling features (including and particularly, firewalling the baseband from the rest of the system).
Yeah, It's definitely on my radar. I had an N900 and it was awesome. The only reason why I ditched it was that it didn't work on all the spectrums I needed.
The Neo900 looks awesome, and has the upgraded radio. It really does tick all boxes, except the one that says "in stock" :)
I've looked at Replicant too, however I'm really not an Android fan. I long for a Debian phone... the Ubuntu phone does looks interesting too. But FWIW - stock Debian or die.
Freedom and security on mobile devices is arguably more important than workstations, given the amount of information they manage.
I'm amazed there aren't large organizations interested in funding a Debian-on-free-hardware phone.
Good question! I believe that CPU microcode is still unavailable (though I could, and would love to be, wrong. I'll check with Francis).
The FSF's RYF certification (which Gluglug machines have) makes an exception for "software delivered inside auxiliary and low-level processors and FPGAs, within which software installation is not intended after the user obtains the product. This can include, for instance, microcode inside a processor,"[0]
Additionally, I known that the firmware on the SSD (if you chose that option) is proprietary--it is covered under the exception as well (as an "auxiliary processor").
However, the criteria goes on to say that if a Free firmware for any exempted components becomes available, the product must "adopt it within a reasonable period of time." (or have their certification withdrawn).
The CPU certainly does have at least patchable control store.
Even ignoring that, there are certainly at least two significant binary blobs still remaining: EC firmware and microcode for e1000.
First probably has no major security implications, second is slightly weird as there were security-significant erratas in e1000 that were fixed by microcode update, but on the other hand it's probably more of "random undocumented configuration options" than "executable code" (for some reason Intel tends to call any random binary blob "microcode").
On a stock kernel you can update it. On the kernel this machine ships with, they removed that, along with most other facilities to upload binary firmware, for things like wifi cards, video cards, etc.
Which reminds me that I don't see the point of skipping microcode updates when you are running non-free microcode when you are resetting the CPU. This is one reason why practically I think being able to modify the BIOS in the first place is more important.
Most of these types of systems will patch out any sort of microcode updating interfaces to keep you from performing such actions. Can't have the user decide to flash the firmware, now can we?
Am I being too cynical when I say that the laptop you and I use will never be a 100% totally free. There will always be a blob on some sort in there. The most obvious one is the processor. With my naive knowledge of processor implementation, it's functionality is implemented in code more primitive than microcode. Is the source for this available for inspection?
There will be plenty of other devices in the laptop where its implementation domain is blurred.
Having said that, the Gluglug is a shuffle in the right direction.
I always thought that it would be nice to handle your essential private keys (PGP master key, large bitcoin wallet) on a fully free hardware. Of course free software still can be compromised but at least you could be certain to a degree that your hardware doesn't come compromised out of the box (sure, you have to trust your "fully free hardware supplier" here).
This looks good, but it's a bit pricey for my tastes and that hardware is an unreliable brick of a machine.
I'd love to see someone hit the market with a working Linux-ready netbook. I used to have an Acer One cobbled into that state, but the lack of direct support from the vendor meant that things like power management and sleep-on-lid-close never quite worked right. It's the little things you miss when you go off the vendor's book on operating system choice.
On the other hand, the stock Vista on that machine was dog-slow. Being able to strip down the crap from the UI and run much faster was a Godsend for hacking around on projects while away from the desk.
Anybody could intercept the connection with an unsigned connection. At least they could have used HSTS so that only the first visit is vulnerable. If you have a signed cert the NSA has to work harder; there are more logs and it's more riskier for them.
It's a refurbished Lenovo ThinkPad X200 (2008, I believe). So yeah, it looks kinda dated. And yeah, ThinkPads tend to look blocky; I don't think that's news to anyone.
