Canadian goods are often still marked up, I find, even though the Canadian dollar reached parity with the US dollar a long time ago. Fortunately, it's relatively straightforward to get around that.
Furthermore, you should keep in mind that you're not comparing two machines that are identical spec-wise. The Macbook Air has an i5 processor whereas the XPS 13 has an i7, and memory/storage-wise, the comparable Macbook Air would be the model that sells for $1499 in the US.
In other words, in the US, the XPS 13 is comparable to the Macbook Air that seems to cost $1,699 when I enter in all the upgrades on Apple's website.
I'm suffering through the usual problems of setting up a new linux transportable laptop right now. It is an ASUS R900V, and Ubuntu 12.04 currently has problems with wifi, graphics, and sound! I would love to be able to buy a transportable (large screen 1920x1080, 8GB, fast cpu) for a reasonabl price in Europe, and that is well supported by Ubuntu.
- it must be cheap, $300 is fine, because a Laptop can easily be stolen, or damaged.
- it must be light, less than 2lbs, because I don't own a car, and still want to carry my laptop around
- it must boot Linux from network to boot into backup cycle.
- the hardware must be 100% supported by free drivers.
- it should not bundle a Microsoft tax.
I don't care much about the preinstalled OS. I'll replace it anyway. I see no sense in bundling a spyware Ubuntu, and even less in some cloud service crapware on a developers laptop. I don't know if other developers think similar, but I fear that the hyped Sputnik will just give Dell the excuse to abandon Linux laptops again, because of low sales.
The class of $300, 2lbs Linux laptops created fear for Intel and Microsoft instead. Thats why they are no longer sold.
> The class of $300, 2lbs Linux laptops created fear for Intel and Microsoft instead. Thats why they are no longer sold.
No, the reason they generally aren't sold is because consumers don't want them -- most consumers would rather have a tablet, as evidenced by sales. It's not some massive conspiracy, it's just that your needs are not the needs of most consumers.
oops "boot into backup cycle" is my method of installing new systems. I create them in a chroot/lxc on a NFS server, and boot the real system or laptop by PXE, rsync it on the disk and make it bootable. The real system will also rsync itself to this master, and the master is doing incremental backup.
So every system has a backup before it even exists on its own disk, and installation is just a normal restore of a backup.
Also on Chromebook, specs tells some funny CPU, and only 2+16GB of SSD. Thats much lower than an old eeepc1000. Especially as it likely ships some funny crippleware, and not a minimal Debian, Arch or Gentoo.
If by "custom" you mean rebranded consumer hardware coming with some Linux distribution I'm never going to use preinstalled, the answer is likely to be "no."
Clarification added later that I never anticipated to be needed: my point here is that there are more significant departures from mainstream needed to call something a "custom Linux laptop for developers": most likely, significant hardware changes.
And those bits on the HDD sure make it harder to install a different OS over the top of it.
I don't understand how you can be snarky about a company that picked an OS (I, for one, am shocked that they picked a popular distribution) and then got a good set of drivers and tools for it, packaged it up and want to support it.
If you want to install something else, install something else. If you don't want to pay the Dell or the Sputnik tax, then go buy something else. I don't get the shitting on Dell for trying to accommodate another market segment.
Maybe so, but that doesn't mean they should not pre-install a free OS. The fact that they official support Ubuntu gives confidence that it will work with (almost) any linux distro (possibly with some work on your end).
It seems to me that Google, with its almost ridiculous surplus of CS PhDs and no legacy desktop OS baggage to drag around is almost perfectly suited to taking Linux and building a power user desktop OS on top of it. If they did, they could build the hardware, too, or at least spec it so that there would be a lot of hardware options that all "just worked" out of the box.
As Apple built a beautiful consumer OS on top of a *nix core, Google could build a power-user OS on top of Linux for the business / developer / smart, skilled people-with-work-to-do market that its PhDs ought to understand more than they understand consumers.
With their money, they could come up with the next gen business platform and build into it the web services orientation they live for and make the hardware / OS / drivers combo to rival Apple, but for influencers instead of the consumers. (They have enough money to buy nvidia if they can't get Linux drivers.)
Who else is better positioned to produce the next gen power-user platform as Windows fades and Apple makes it clear they are a consumer electronics company?
Exactly. Each of those things is a mini-version of something that has been a business standard. I'm talking about creating the next generation of the full-scale items but targeted, not at consumers, but at the power users. Apple gets the consumer market for whom being hip and being simple (a natural match) are priorities. Microsoft tries to chase Apple for the consumer market as it lets its Windows/Office franchise gradually obsolesce. Oracle is all enterprise back office. Samsung is all consumer electronics, like Apple but without the software. So why couldn't Google go for the full-scale power user market that, as Microsoft runs off after Apple, is increasingly uncontested? Say, Linux laptops as well-behaved as Apple (graphics, media, sleep/wake, battery life, networking) but optimized for power and customization, not "we know better than you what you need" simplicity.
They should first build an IDE for Android rather than keep piggybacking forever on Eclipse which is not very lightweight or suitable for a specific purpose, let's say.
Why? Eclipse is just reusable UI widgets and a Java editor. That's basically what you want for writing an "Android IDE", whatever that might be. (I personally just use Emacs.)
It never works seamlessly. I would love to do away with all the clinks and cracks which becomes an integral part of it. Becomes more painful when you have to deal with the "whole code" (say guys at those OEMs and those indie ROMs).
I fail to see how throwing away Eclipse would solve this problem. What you probably want is more engineer time thrown at Android tools, which is something I can certainly agree with.
Eclipse is a popular Java IDE (with plugins for other languages). Android programs are java with an XML file describing the UI. No need to re-invent the wheel here. If you have another IDE you want to use, see if there is an android plugin for it, or right the XML files yourself and use any java IDE, or use a plaintext editor and curse Google for making Android run Java.
Do you know why all the other text editors were built even after the first one was already in use? Try finding answer to it and I am sure you'll find the answer to what is reinventing the wheel and why is it very different from having an IDE that specifically caters to the need of Android development, with total focus and integration, and is supported/promoted/maintained by Google - somewhat the way Xcode is done or hopefully even better.
They already built a new user space for Android. Sometimes wheels need re-inventing, but that does not mean you should abandon all established ecosystems for the sake of inventing the wheel yourself.
> Canonical added the idea of connecting the laptop to the cloud, so that users could develop on LXC containers, replicate the environment on the actual client, and then jettison it to the cloud via the JuJu service deployment and orchestration framework.
This is a very nice idea (although with mobile share plans, we may see 4G support gain popularity in laptops, making this less of a factor). However, I feel that the hardware was neglected to the point that software gimmicks can't make up for it.
I wrote about my experiences with the ThinkPad X1 Carbon on HN a week back[0], and the hardware simply blows the XPS 13 out of the water. The fact that installing Xubuntu (which is much more pleasant for devs and power users than Ubuntu) on an X1 Carbon is laughably easy doesn't help Dell's case.
44 comments
[ 2.5 ms ] story [ 97.1 ms ] thread* Dell Price $1449: http://www.dell.com/us/soho/p/xps-13-linux/pd
But for the rest of us, for instance in Canada (where I'm automatically redirected):
* Dell Price $2,344: http://www.dell.com/ca/enterprise/p/xps-13-linux/pd
On the other hand, the MBA averages (both Canada/US) at ~1199$.
I don't feel embarrassed.
Furthermore, you should keep in mind that you're not comparing two machines that are identical spec-wise. The Macbook Air has an i5 processor whereas the XPS 13 has an i7, and memory/storage-wise, the comparable Macbook Air would be the model that sells for $1499 in the US.
In other words, in the US, the XPS 13 is comparable to the Macbook Air that seems to cost $1,699 when I enter in all the upgrades on Apple's website.
my requirements for a Laptop are:
- it must be cheap, $300 is fine, because a Laptop can easily be stolen, or damaged.
- it must be light, less than 2lbs, because I don't own a car, and still want to carry my laptop around
- it must boot Linux from network to boot into backup cycle.
- the hardware must be 100% supported by free drivers.
- it should not bundle a Microsoft tax.
I don't care much about the preinstalled OS. I'll replace it anyway. I see no sense in bundling a spyware Ubuntu, and even less in some cloud service crapware on a developers laptop. I don't know if other developers think similar, but I fear that the hyped Sputnik will just give Dell the excuse to abandon Linux laptops again, because of low sales.
The class of $300, 2lbs Linux laptops created fear for Intel and Microsoft instead. Thats why they are no longer sold.
No, the reason they generally aren't sold is because consumers don't want them -- most consumers would rather have a tablet, as evidenced by sales. It's not some massive conspiracy, it's just that your needs are not the needs of most consumers.
So every system has a backup before it even exists on its own disk, and installation is just a normal restore of a backup.
Also on Chromebook, specs tells some funny CPU, and only 2+16GB of SSD. Thats much lower than an old eeepc1000. Especially as it likely ships some funny crippleware, and not a minimal Debian, Arch or Gentoo.
Clarification added later that I never anticipated to be needed: my point here is that there are more significant departures from mainstream needed to call something a "custom Linux laptop for developers": most likely, significant hardware changes.
I don't understand how you can be snarky about a company that picked an OS (I, for one, am shocked that they picked a popular distribution) and then got a good set of drivers and tools for it, packaged it up and want to support it.
If you want to install something else, install something else. If you don't want to pay the Dell or the Sputnik tax, then go buy something else. I don't get the shitting on Dell for trying to accommodate another market segment.
As Apple built a beautiful consumer OS on top of a *nix core, Google could build a power-user OS on top of Linux for the business / developer / smart, skilled people-with-work-to-do market that its PhDs ought to understand more than they understand consumers.
With their money, they could come up with the next gen business platform and build into it the web services orientation they live for and make the hardware / OS / drivers combo to rival Apple, but for influencers instead of the consumers. (They have enough money to buy nvidia if they can't get Linux drivers.)
Who else is better positioned to produce the next gen power-user platform as Windows fades and Apple makes it clear they are a consumer electronics company?
Well, not exclusively, if I recall correctly, they get to choose Goobuntu, Windows, or OS X.
Also, I believe they stopped offering Windows a couple of years ago.
There really isn't a reason for them to release it externally. Just use Ubuntu LTS if you want what they're using.
This is a very nice idea (although with mobile share plans, we may see 4G support gain popularity in laptops, making this less of a factor). However, I feel that the hardware was neglected to the point that software gimmicks can't make up for it.
I wrote about my experiences with the ThinkPad X1 Carbon on HN a week back[0], and the hardware simply blows the XPS 13 out of the water. The fact that installing Xubuntu (which is much more pleasant for devs and power users than Ubuntu) on an X1 Carbon is laughably easy doesn't help Dell's case.
0: http://news.ycombinator.com/item?id=4848375