Claes Wallin (韋嘉誠) [email protected]
Hong Kong
Linux native since 1995. Working in Ruby, shell scripts, whatever gets the page up. Solving yesterday's problems tomorrow.
-
What You Can Do To Increase Cell Phone Security
Kete Foy at 2016-06-06T04:28:08Z via AndStatus To: Public
“To avoid bugging, keep your phone in your control as much as possible. If you are forgetful, try duct taping it to your hand. ”
http://www.howtovanish.com/2010/01/cell-phone-security-mobile-phone-taps/
“Actually, the problem revolves around the entire concept of the "Smart Phone". First off, most people DON'T NEED ONE. They're nice, and convenient, I grant you, but they're TOO SMART. You don't NEED the equivalent of a 5 year old desktop computer in your pocket to make a few calls. You don't NEED internet access in your phone. You don't NEED the $100+ per month charges. You don't NEED a device with the ability to monitor your every move and word. You don't NEED the $2500+ total cost of that fancy phone over the 2 year contract period. I would suggest we return to the old, reliable feature-phone, perhaps with a QWERTY keyboard for texting, and maybe a camera, but with a removable battery. None of us, except maybe some corporate customers, have any real need for a "smart phone" in any form.”
[email protected] , Icaro Perseo like this.
Claes Wallin (韋嘉誠) shared this.
I agree! Simple solutions are always the best choice. ^.^
Icaro Perseo at 2016-06-04T20:23:43Z
Douglas Perkins likes this.
The “you don't need X” argument is very weak. We don't need most things we buy … what of it? :-)
Indeed. No, I don't need internet access when I'm not at home, actually humanity doesn't *need* internet access at all, we have been doing just fine for ages with no internet at all, haven't we?
Except, having internet always available is very convenient and enables a number of things that wouln't be possible otherwise.
Just one example out of many: when buying used hardware being able to check for compatibility with free software on the spot is much better than having to travel with a printout of a full list of compatible/non compatible devices (and hope that the ones actually available are included in the list)
Admittedly, I wouldn't spend those prices to have it: in my country you can easily buy a few GB of traffic (and some calls and texts) for about than 10 EUR/month, and AFAIK everywhere you can buy low-range android smart phones for about 100 EUR or so, and they allow more or less the same basic smartphone things that a 500+ EUR phone does (sometimes not at the same times, sometimes in a clunkier way, surely in a less shiny package, but the basics are always there).
BTW, I've never actually bought such an Android device, but I have a ZTE Open phone with Firefox OS, which is definitely low-range, definitely a smartphone and comes with a removable battery, so you can have it and internet at the same time, no need to use a feature phone for that.
Elena ``of Valhalla'' at 2016-06-05T13:23:26Z
Douglas Perkins , Claes Wallin (韋嘉誠) , Kete Foy like this.
-
Elena ``of Valhalla'' at 2016-06-06T04:17:23Z via AndStatus To: Public
Evil 32: Check Your GPG FingerprintsIt takes 4 seconds to generate a colliding 32bit key id on a GPU (using scallion). Key servers do little verification of uploaded keys and allow keys with colliding 32bit ids. Further, GPG uses 32bit key ids throughout its interface and does not warn you when an operation might apply to multiple keys. Key servers do not use transport encryption (e....
Sadly, one of the main tools used to analyze Web of Trust data, wotsap, is still using 32 bit key ids in its data files, and is mostly abandoned upstream, so there are little real hopes to see it fixed. Wotsap data is also used by the PGP pathfinder & key statistics website, which is thus vulnerable to a number of attacs.
The workaround is to manually verify the paths shown by wotsap usinggpg --check-sigs , which should be done anyway, since wotsap data comes from an untrusted source (cryptographically speaking), but AFAIK is still not done automatically by any tool.
Thanks to @Enrico Zini for the link.Claes Wallin (韋嘉誠) shared this.
-
Taiwan
Douglas Perkins at 2016-06-04T12:32:32Z via AndStatus To: Public
Anyone been to Taiwan? I'm thinking of going this August. Expect it to be hot, except up in the mountains and maybe even then. But still should be an exciting trip. Thinking of maybe three days in Taipei and then some coast and mountains on the eastern side of the island for another four?
My friend lives near Hong Kong, but he says it's way too hot there in the summer, so I suppose the winter or spring would be better. Leaves the summer for Taiwan.
Claes Wallin (韋嘉誠) shared this.
I lived in Hong Kong for a couple of years and visited Taipei over one weekend. Hong Kong is best around May and August. 30 degrees Celsius with 90% humidity is not enjoyable. When in Taipei, don't miss the volcanic crater in the mountains a bit outside the city.Claes Wallin (韋嘉誠) at 2016-06-04T12:36:15Z
Douglas Perkins likes this.
To clarify: August in HK isn't too bad. June and July is the worst part. And I hate HK winters. I prefer Swedish winter with snow and with decent indoor temperature to HK winter with 10 degrees C outside, 10 degrees indoors and still 75% humidity. Freezes me to the bone. If you have the choice, stay out of HK Nov-Feb.Claes Wallin (韋嘉誠) at 2016-06-04T12:40:22Z
Douglas Perkins likes this.
-
We Distribute at 2016-06-04T11:03:58Z via AndStatus To: Public
What's going on with OwnCloud?
The recent announcement of the OwnCloud Foundation followed by the announcement of fork of OwnCloud is bound to raise more than a few eyebrows. What happened? How did things get to this point?
Catching the Next Cloud
Earlier today, Frank Karlitschek, co-founder of #OwnCloud, made an announcement about the formation of a new project. Five weeks prior, he had left the company that he had helped start, and some of the core development team have followed him to their new project, NextCloud.
Frank describes a vision for what he wants out of this new project and its respective community: a project that does development out in the open, without any contributor agreements or dual-licensing schemes, with a new trademark held by an independent foundation. A better company culture where shares are equally held by employees, with the hand of development decisions guided by the community-at-large rather than a disconnected small team.
To paraphrase Frank, the best parts of OwnCloud are to be extended to fully support what the community has been asking for: better apps, with features that previously only existed for enterprise customers. In his own words:We will release a drop in replacement for ownCloud in a few weeks so that users and customers can easily upgrade to Nextcloud to benefit from the new bugfix and security improvements and features. Nextcloud GmbH will provide free support for all current ownCloud customers to simplify a transition.
The OwnCloud Foundation"One of Frank’s criticisms concerned the need to strengthen the Community. In this regard, we have been working on the creation of the ownCloud Foundation, the formation of which we announced earlier this week."
Recently, the OwnCloud project announced that it was setting up a foundation. While this is a welcome event for any Free Software project, it immediately raised eyebrows. Some of the main criticisms were focused on the nature of the board itself (only 2 seats are elected by the community - the other four are coporate), whereas others have pointed out that only OwnCloud Inc itself appears to have any sponsorship in the foundation.
Meanwhile, a handful of developers of the OwnCloud platform announce that they are leaving their projects. Their announcements can be found on Planet KDE. The general theme appears to be that OwnCloud is putting its corporate needs above its community needs.
The OwnCloud Foundation's had this to say on the matter:Today’s announcement by our former colleague Frank Karlitschek, that he intends to launch a competitive product to ownCloud into the market using recently poached developers, has both surprised us and – admittedly – disappointed us. In the past, Frank has made a wealth of contributions to the development of the ownCloud Community Edition. With today’s announcement, he is no longer related to the ownCloud project and has started a competing community.
This ends on a particularly strange note, as OwnCloud finally concludes that it has to close its doors after their main lenders cancelled their credit. 8 employee contracts were terminated in the process, and the company will be forced to close its doors. Paradoxically "The ownCloud GmbH is not directly affected by this and the growth of the ownCloud Foundation will remain a key priority."
#NextCloudClaes Wallin (韋嘉誠) , Jorge , sazius , gledof and 2 others like this.
Stephen Michael Kellat , mnd , Stephen Sekula , Blaise Alleyne and 6 others shared this.
This ends on a particularly strange note, as OwnCloud finally concludes that it has to close its doors after their main lenders cancelled their credit. 8 employee contracts were terminated in the process, and the company will be forced to close its doors. Paradoxically "The ownCloud GmbH is not directly affected by this and the growth of the ownCloud Foundation will remain a key priority."
I'm confused. What closed its doors? I would've thought the business, but the next sentence says the GmbH is not affected. Anyone know?
I think I've read somewhere, that the US branch "Inc." has to close down, while the "GmbH" which is the German branch can continue. But I don't have a link to prove.Tobias Diekershoff at 2016-06-03T16:19:55Z
Claes Wallin (韋嘉誠) , Mike Linksvayer like this.
>> Mike Linksvayer:
“This ends on a particularly strange note, as OwnCloud finally concludes that it has to close its doors after their main lenders cancelled their credit. 8 employee contracts were terminated in the process, and the company will be forced to close its doors. Paradoxically "The ownCloud GmbH is not directly affected by this and the growth of the ownCloud Foundation will remain a key priority."
I'm confused. What closed its doors? I would've thought the business, but the next sentence says the GmbH is not affected. Anyone know?”The developers themselves and the promise of their future intellectual properties was the asset or other collateral used to secure the debt?
Stephen Michael Kellat at 2016-06-05T19:41:54Z
Alex Jordan likes this.
-
Christopher Allan Webber at 2016-06-03T02:15:25Z via identi.ca To: Public , Christopher Allan Webber CC: Followers
It turns out collapsing on the floor for a while was the right idea after all.
Nathan Willis , Douglas Perkins , Claes Wallin (韋嘉誠) , Alex Jordan and 3 others like this.
Douglas Perkins , Claes Wallin (韋嘉誠) shared this.
-
Sean Tilley at 2016-06-03T02:14:00Z via identi.ca To: Public , Sean Tilley
> Sounds a bit like you have Nothing To Hide.
Nah, I've always thought of that argument as a bad one. Of course people have things to hide. It's just that I'm not that worried about music listening habits. Can someone scrape that data and monetize it? Probably. But I'm putting it in a public place willingly.Douglas Perkins , Claes Wallin (韋嘉誠) , Alex Jordan like this.
Claes Wallin (韋嘉誠) shared this.
Exactly. I have chosen to air my views. Anyone who tries to figure out who I am will do it. Adding my music taste to the puzzle doesn't add much. Doesn't mean I think privacy isn't important, I totally understand that there are people who cannot make the same choices I made.
Claes Wallin (韋嘉誠) at 2016-06-03T02:14:57Z
Sean Tilley likes this.
-
Kevin Everets at 2016-06-03T02:12:38Z via identi.ca To: Public , Aaron Gibson , Freemor
@[email protected] @[email protected] I'll keep an eye on it, too, but being that the main anti-Distribution actors are part of NextCloud, I doubt I'll be using it any time soon. I really want a syncing solution that has a proper apt-get or "guix package" solution. Got so tired of ownCloud breakage on every update.Alex Jordan , Douglas Perkins , Claes Wallin (韋嘉誠) like this.
Claes Wallin (韋嘉誠) shared this.
-
2016-06-03T01:49:02Z via Microca.st Web To: Public CC: Followers
I have come to think that federations should be at best frontends to decentralized systems. The concrete example is various web wallets for crypto currencies. I don't know enough about Red Matrix, but I suppose from what I know that it is an example as well.
Identity is the key, as Moxie points out. You can start out on a federated service, which really means a super node in the P2P network, but then you need to be able to take your identity outside without losing your connections.
So, you have a decentralized system at the bottom. Decentralized as in truly P2P. Anyone could participate by just having a client that interacts with the network, without relying on any one particular server.
But people aren't going to do that ... you know, except the ones who are. So you have the situation with identi.ca vs OStatus, but with a third level. There's the big centralized federal entities vs the small self-hosted-with-friends or -with-orgs entities vs the P2P decentralized I'm-just-on-a-mobile-app entities. And they are all talking.
Through the web and the big entities you discover the network. Through the big or small entities you start participating, and with your P2P client you find emancipation and the network lives on regardless of what happens to the big players. But the big players are needed. How many of the people on Facebook know about Twister or BitMessage or whatever? But if the "Facebook" of a P2P+Federation network, with distributed identity, would start censoring, people would really quickly find another federated peer to start publishing on, and if they were worried enough, they would finally take the step off the web and run a desktop or mobile client.
I think there's a lot of amazing P2P social network apps out there, but if they are going to go anywhere in a big way, they need to be on DuckDuckGo, on Google, on Bing, on Yandex, on Baidu. We need searchable frontends and public content, otherwise it's just a niche for cryptoheads. But then we need inroads to heavier decentralization. Pump.io lets you run a client, but you're still reliant on a server being up.
I want git over secure-scuttlebutt, forums over IPFS and SAFE, IM over Ring or Tox or whatever, but I want it discoverable for civilians. And easy onramps toward more decentralization, with clear benefits.
It's not easy, because then people would have done it already.
So, W3C WG Social, awesome that you exist and I hope ActivityStreams 2.0 becomes a thing because we need a vocabulary either way. The messaging revolution will need solid work to stand on, but somebody had better start working on the BitCoin of messaging, rather that a more open, more free, more accessible and sensible SWIFT network.
Who am I? I am nobody, I didn't even finish the tent poles of my social network vaporware. But I can dream, can't I?
Alex Jordan , Charles ☕ Stanhope , Douglas Perkins like this.
Alex Jordan , Alex Jordan shared this.
-
$LOCPATH is broken on Ubuntu 16.04/xenial. Here's how you fix it!
2016-06-02T16:21:55Z via Puma To: Public
Ten years ago, a user on LWN noted that a simple tool like
catdoes a whole bunch of file lookups on Ubuntu because of the way locales are set up in glibc.Ulrich Drepper, famous/infamous then-maintainer of glibc, commented within hours that anyone not using
/usr/lib/locale/locale-archive, e.g. Ubuntu, is either “mistaken”, “ancient” or “cannot even keep [their] machine running correctly”.The reason Ubuntu did it this way was inherited from Debian: With
local-archivepresent instead of the olden_USet al folders, usingLOCPATHto provide several locations for locales stops working. It's either-or. You setLOCPATH, glibc ignores the cache.Fast-forward to 2016, this has now been “fixed”. So if you want to run Nix on Ubuntu, you have two choices: Break locales in Nix, or break them in Ubuntu.
They didn't make it configurable either. They just broke legacy and moved on.
Well. This is Debian-based, so there's a solution of course. You can replace
/usr/sbin/locale-gen, make it run the originallocale-genwith--no-archive, and you can make your change locale-package-update-proof usingdpkg-divert!Solution (hashify.me mega URL!)
TODO: Fork http://hashify.me/ to use fragment instead of sending the full URL to the server, and to use all relative hrefs so it doesn't have to be employed on the root of its domain. Then deploy on freenet and ipfs/neocities.
Alex Jordan likes this.
Btw, guix is not affected, because their glibc uses GUIX_LOCPATH instead to avoid this kind of collision.
Claes Wallin (韋嘉誠) at 2016-06-02T16:23:36Z
X11R5 likes this.
And, so far, Debian has not been "fixed", so Nix+Debian works without locale issues.
-
at 2016-06-02T05:06:51Z via AndStatus To: Public
Alex Jordan , Claes Wallin (韋嘉誠) , Efraim Flashner , Kete Foy like this.
Claes Wallin (韋嘉誠) shared this.
-
CPython going github
2016-06-01T15:22:27Z via Puma To: Public
core-workflow: We will be moving to GitHub
This is from January. I don't follow Python core development, and only stumbled over this because I'm learning about Zuul, a post-review branch merger validator thing with parallelizable speculative branch merging as its main feature. It's pretty cool, check it out.
Once, there were PEP-0474 and PEP-0462, two suggestions for moving CPython development to Zuul and Kallithea. I'm guessing Conservancy and bkuhn had a hand in this, as Bradley has been talking about wanting to nudge CPython to use Kallithea, and his name is even mentioned in one of the proposals.
Then there was PEP-0507, and the other two were withdrawn. This proposal had GitLab and git instead of Kallithea and Mercurial, and I buy its rationale. While it would be awesome for Python to be developed using Python tools, the network effect for git is huge, even in Python circles, and back when I played with it, Kallithea was far behind GitLab in functionality.
While using Kallithea for Python development would have given it visibility and probably would have meant a push for improved functionality, I get that CPython developers might not want to divert effort away from core development. And it's all free software anyway!
So PEP-0507 made a lot of sense. But back in January that was rejected too.
From that email linked at the top of this post:
“No major distinguishing features between GitHub or GitLab”
Wow. GitLab being free software is a huge feature.
“and no, being open source is not enough of a feature; as I said when I started this process, being open source would help break ties or minor lead of one tool but not be a deciding factor”
Oh. But even with this reasoning, GitLab would have had the edge. As the first quote said, there are no major differences.
“Familiarity amongst core devs – and external contributors – with GitHub”
I guess this is the killer. GitHub network effects. Or maybe “Guido prefers GitHub” is the actual killer, even though it's buried as the #3 reason.
Really, there are no major differences between the two platforms, except what I consider the the killer feature of GitLab being free software. If you have used GitHub, you understand GitLab. Especially as they're not going to use issues and wiki, those are staying on Roundup and MoinMoin.
In my view, the Python Software Foundation missed the chance to promote free software in a big way, at very little cost. This move is going to be a lot of work regardless if they had moved to GitHub or GitLab, and once it's done it's going to take a long time before anyone will be prepared to move again. Python is stuck on a proprietary source code host for the foreseeable future.
Maybe it was the right choice for them, maybe it does lower thresholds and bring in some of those drive-by contributions that we all want every free software project to be able to get. Time will tell.
I think a project as big as Python is probably intimidating enough to contribute to that the hosting platform is not a big part of the filter, and bringing such a big project over to the strongest free software GitHub challenger would have been a big deal in terms of opening people's eyes to the fact that GitHub is not the only way to do git.
But in the end, like I said, at least they are not using the big vendor lock-in component – issues. It's just git for now. And once peer-to-peer git becomes a thing, whether it's git over secure-scuttlebutt, something layered over IPFS, or something we haven't yet seen, maybe it won't be as difficult as I think to move the community over to that new thing and finally get rid of that big proprietary Achilles heel of the free software world, GitHub.
jvalleroy , Alex Jordan , sazius , Charles ☕ Stanhope and 1 others like this.
@arnebab quotes several community reactions to the Python GitHub move:
http://www.draketo.de/py2guile/free
In particular, Brett Cannon's background and rationale post on the Python->GitHub move is a relevant read:
http://www.snarky.ca/the-history-behind-the-decision-to-move-python-to-github
-
2016-05-25T06:23:25Z via Microca.st Web To: Public CC: Followers
Got reading glasses.
My screen is pretty dirty. I never noticed.
sazius likes this.
My glasses are pretty dirty and I seldom notice that :-)
-
Freemor at 2016-05-24T13:40:17Z via identi.ca To: Public , Alex Jordan CC: Followers
@[email protected] but Moxie isn't saying federation is difficult. He's saying centralization is the only way forward (Or at least that is how I hear it). And while doing so failing to point out the failings of a centralized system (which are many and well understood).
I probably would not have such strong reactions if I felt he was trying to open a involved discussion on the subject but sadly his posting read like statements of proven fact rather then an opinion.
If federation is dead and unworkable then so is the entire internet because it is built on a massive number of federated systems.
Also I suspect that some of the strong reactions come from the fact that the federated/centralized (also open standard vs proprietary) discussion encompasses much more the a discussion about design but also touches on some fairly emotionally ladened ideologies about what the Internet should be. Free and open where everyone can run and control their servers or centralized where only the major outfits can play.Alex Jordan , Claes Wallin (韋嘉誠) like this.
Claes Wallin (韋嘉誠) shared this.
He is disillusioned with not achieving the things he wanted to achieve and is rationalizing his way to cope with it.If federation is dead and unworkable then so is the entire internet because it is built on a massive number of federated systems.
Well, he is basically saying that all 80s technology on top of TCP/IP is dead.
He is doing a reverse Batman. If he can't be the hero internet users need, he can be the hero they deserve.
Claes Wallin (韋嘉誠) at 2016-05-24T13:42:30Z
Freemor , Charles ☕ Stanhope , Alex Jordan , bthall like this.
-
Freemor at 2016-05-24T13:30:35Z via identi.ca To: Public , Claes Wallin (韋嘉誠) CC: Followers
@[email protected] Honesty the subtext I'm hearing in these posts is "If you want to have loads of captive users and make a shit load of money off exploiting them..."bthall likes this.
Claes Wallin (韋嘉誠) shared this.
That's not what I'm reading at all. I'm reading that he's tired of not making popular software.
It's Free Software vs Open Source deja vu all over again.
There are reasons he got away from free software.
Updates. Security can be ensured (ethically) through patches and that patches go through is of prime importance and time not lost is time gained. Google Play comes to play. He talked earlier that there be a system which nags the hell out of the user. Build alternative.
While free software is great. The idea and people behind are awesome and cool, we as nerds try to control by not controlling, we set the system which isn't welcome to new changes >> federation. We free software people tend to resist change (which is natural for introverts, geeks, hackers), which is natural (change is). While pump and discourse are good enough for us, name one who isn't nerd, geek, tinker kid or hacker who is using diaspora, pump.io, or status.net
Federation is great but securing a federated system isn't easy. Where do trust come from? Well in theory it has to be one and only that who can authenticate, key servers into play >> centralized.
There is bitmessage who is using that?
It takes awful lot of time to connect and downloading data and subsequently the real messages I am waiting for. Maybe developers aren't doing it better. Maybe.
Why libresignal people insist on using signal's servers? Why moxie denying access to third party apps (libresignal) is looked down upon?
I think he is right, there is signal which offers security they claim. There are others who if he supports bears his name, a layman like me be like they must be same which they aren't. They didn't make same decisions moxie did. The signal he represents should be the one he put his best work in. The code he released under GPL.
There are us who think there isn't no one who be dictating what I name my project, Right. Then him >> I don't want it as law but as goodwill or norm, for people like me i.e layman would know signal is what he is saying is than derivatives. For someone like say me libresignal looks > Signal plus free software, all good there. I be using libresignal than signal.
Who knows people who don't know much about computers or aren't from nerd, geek, hacker culture, who are converse via xmpp? All those running their emails, xmpp, pump, status or diaspora or discourse servers are almost entirely those for whom computers is where fun starts. Yeah there are people who have got started hosting their wordpress site themselves, but they are there because shared hosts like bluehost, dreamhost made it possible for them via their control panels. Again something moxie is doing bringing benefits of technology to those who aren't fluent. Read his GPG post.
There are things I wish moxie hadn't done like say being too critical of project naming or going closed source with redphone. But I am not dejected by him. I can understand/sympathize where he is standing or his position on certain things.
Yes, I have been inspired by moxie a lot, him being a sailor in same class as Moitessier, anarchist, his work on cryptography protocols. Signal. sslstrip and like. Hell, I wanted to be like him.
Call me an apologist or anything. This is something I think is how it is. However I would like to know what [email protected] thinks.
testbeta at 2016-05-24T20:35:13Z
Claes Wallin (韋嘉誠) likes this.
Federated systems are still centralized at end points, who here are using identi.ca, quitter.se, microca.st or gmail, yahoo, apple's or likewise. My point is people put trust in major players that trust is justified. In the end major players have most of the userbase. Federation is still in infancy. -
2016-05-24T07:31:22Z via Microca.st Web To: Public CC: Followers
The European Court of Justice gets it right!Computer code itself can be copyrighted, but functional characteristics—such as data formats and function names—cannot be. "To accept that the functionality of a computer program can be protected by copyright would amount to making it possible to monopolise ideas, to the detriment of technological progress and industrial development," the court stated.
"The purchaser of a software licence has the right to observe, study, or test the functioning of that software in order to determine the ideas and principles which underlie any element of the program. Any contractual provisions contrary to that right are null and void," the court ruled.
Ars Technica: EU’s top court: APIs can’t be copyrighted, would “monopolise ideas”
/by Timothy B. Lee
Charles ☕ Stanhope , Efraim Flashner like this.
-
2016-05-21T03:17:53Z via Microca.st Web To: Public CC: Followers
Says Moxie Marlinspike:I understand that federation and defined protocols that third parties can develop clients for are great and important ideas, but unfortunately they no longer have a place in the modern world.
Ouch.
https://github.com/LibreSignal/LibreSignal/issues/37#issuecomment-217339450
/via https://quitter.se/notice/5688952
This relates to that whole "federation is dead" post also by Moxie:
https://whispersystems.org/blog/the-ecosystem-is-moving/
Alex Jordan likes this.
jrobb , Alex Jordan shared this.
Show all 10 replies
To the apologists, be careful. Moxie's words were strong and clear. Also, at odds with reality (see for example all of the examples others mentioned above).
It is better to say “Moxie is wrong about this.” than to pretend he did not say incorrect things. We all get it wrong from time to time.
I read it as saying "centralization is the only way forward because federation is so difficult." A statement which overall I disagree with but certain parts of which have some validity.
Alex Jordan at 2016-05-23T03:58:19Z
Claes Wallin (韋嘉誠) likes this.
Douglas, was that comment aimed at me? If so, I'm definitely not being an apologist for Moxie. I think his arguments have some merit, but as I said, I think his conclusion is incorrect and that his argument as a whole doesn't hold water.
Alex Jordan at 2016-05-23T03:59:58Z
Claes Wallin (韋嘉誠) likes this.
Alex, not aimed at you specifically. A general note of caution about how not to frame things.
On re-reading the comments, everyone (including you) here was fairly precise and accurate, which is pretty cool (and rare).
Douglas Perkins at 2016-05-23T07:29:27Z
Claes Wallin (韋嘉誠) , Alex Jordan like this.
-
2016-05-19T14:00:56Z via Microca.st Web To: Public CC: Followers
How do you easiest enable a Windows machine to be seen over mDNS? Install iTunes.
It's like the two evils partially cancel each other.
it's not so much iTunes as the fact that it installs bonjour (Apple's mdns responder). installing iCloud for windows I suspect would do the same thing.
I would have guessed installing iCloud would imply installing iTunes. iTunes is basically "OSX for Windows".
If iCloud doesn't require iTunes, it probably doesn't install mDNS either, it's not like it's required to push things to Apple's servers. Or does it involve some local discovery thing as well?
-
2016-05-18T19:18:10Z via Microca.st Web To: Public CC: Followers
"Every time you decide that making users feel stupid is better than fixing your code, you're making an ethical decision."
http://www.joelonsoftware.com/news/fog0000000322.html
Jason Self , Alex Jordan like this.
Alex Jordan shared this.
I have seen too many (usually proprietary) things where this applies. Can't tell you how many people I've told. "No it's not you.. the program is crap"
Freemor at 2016-05-18T20:07:17Z
Claes Wallin (韋嘉誠) likes this.
-
docker-nix
2016-05-18T01:55:45Z via Puma To: Public
I would love to build something like this for guix.
Meanwhile...
-
Claes Wallin (韋嘉誠) favorited an image
-
Claes Wallin (韋嘉誠) favorited Bug fixing
-
Claes Wallin (韋嘉誠) favorited a note
-
Claes Wallin (韋嘉誠) followed an unknown object type: 113612
-
Claes Wallin (韋嘉誠) favorited sweet fiber optics
-
Claes Wallin (韋嘉誠) favorited a note