Reddit: The dark side of open source's powerful tool: Fusion of tracker data and traces from web browsing will soon make privacy a thing of the past
I would like to discuss the fact that open source is not only powerful, but also provides tools which do seriously threaten to dismantle any privacy as we understand it today. Any interaction with big web services like news sites, search engines, social media, email services and online shopping sites is stored in databases. Companies are working full steam to connect and de-anonymize that data. The technologies they use are very much open source - map/reduce, Hadoop, NoSQL databases, parallel computing, and many things more.
And they are increasingly successful as this quote from a fairly technical article shows:
"For instance, one question that is easy for a salesperson to pose is: How many female WWF enthusiasts under the age of 30 visited the Toyota community over the last four days and saw a medium rectangle? (A "medium rectangle" is a standard- sized web ad.) The goal is to provide answers within minutes to such ad hoc questions, and it is not acceptable to refuse to answer questions that were not precomputed"
(cited from page three, fifth paragraph in second column)
In short, the current goal of advertisers it not only to link some data, but to actually track individual people and link their identities to as much data as possible. Here a more complete list about browser fingerprinting techniques:
And this results to activities like this - predicting pregnancy in online buyers:
It is easy to predict that if this continues for a few years, online privacy will be pretty much gone. As the flood of data from online activities such as shopping is every time growing and the "internet of things" as well as, for example, clothes tagged with RFID labels add to this, data items which cannot be linked to any other items connected with you will become fewer and fewer. This can be compared to an information avalanche. For example, if the RFID numbers of a few parts of your clothes are known, they will not only be able to identify you as you walk through a mall, you will also tell every transceiver you walk upon if you have new clothes, possibly even item's you didn't purchase yourself, but were gifted from your new acquaintance which went shopping on Valentine's day. The rising flood of data leads to an avalanche of information about their relationships and connections. This is wonderful and exciting from an scientific and information technology point of view, but horrific if you consider that access to information and the ability to analyse it means power and that power is far from being distributed in an equal way.
I think it would be interesting to understand how computer systems would need to be designed to give users the largest possible amount of control over their data.submitted by greenbyte
[link] [14 comments]
Desperately needing solid-reliable storage at home, for home videos, photos, music, pdfs, and as a storage backend what will become a home lab, I've set to looking to buy a good NAS enclosure. I've spent 3 days looking at options and would like some advice. I've looked at every idea, including building a superweird system based on a parallella. But I need to get serious. I'd like:
- BTRFS support (civil discussion is welcome, but my research so far is telling me I like the way it is heading).
- Potentially install nixos
- Reliable RAID
- Fairly Quiet
- Not too tall. Need to fit on the shelf of a new home office desk.
- Maybe simple graphics, to act as a workstation (until I can get the time to build a 3D workstation)
I don't like:
- Anything too proprietary
- Too much gimmicks and web UI crap. Rockstor is about the most I can handle. I'm a software-only "sysadmin/devop" by day so I prefer to do everything by CLI.
- Not reliable. I don't want to spend my weekends recovering data from a broken tablet or laptop (anymore).
I think my options are:
- Buy a NetGEAR ReadyNAS Duo V2 which has btrfs support built in to ReadyNAS OS6. Pros: ARM, btrfs, pre-built.
- Buy a cheap d-link dns-320 which can be hacked to run mainline debian kernel. Pros: Works out of the box, cheap, can be hacked.
- Buy a quad or octa core ARM mini computer with USB 3.0, plug a USB SATA enclosure, and just back it up regularly. In a way, backups are better than btrfs (you still need backups for accidental deletions and PEBKAC). The newly released Tronsmart Draco AW80 might suit. Pros: ARM, certainly would allow light desktop use. Could use it as a build box for ARM nixos. Cons: probably a lot of work, not many people running btrfs etc.
- Build a box. This seems to be a pretty good option nowadays. Extendable, runs nixos, btrfs, everything I could want it do it. I'm not sure about the whole AMD M1 vs Intel Ivy Bridge? etc etc I'm thinking something like this, but I haven't built systems for my day job in 15+ years, so any help would be appreciated:
- Something else I haven't thought of yet. I'm sure others have gotten a home solution which is better than any of the above.
[link] [22 comments]
Trying to install Ubuntu on second drive. A new SSD initially formatted into 4 NTFS chunks with nothing left over
I think if I knew how to get to the partitioning menu where it asks for a mount point maybe problem solved, except don't know if have to do that seciton as ext4.
See photo, Ubuntu trial from dvd. Picking one of the big sections and clicking install gives
"Root file system not defined. Please correct this from the partitioning menu."
It allows "New Partition Table" if you have the physical drive selected in the window. I did that, so now I have a solid green free space bar across the top for 250 GB
(bad pic from phone, I'm guessing print screen probably will not work running this way)submitted by chazzacct
[link] [5 comments]
Reddit: Looking for old post: Operating system utilizing VM software to run apps in separate environments.
I believe this was a ex-hacker turned security consultant.
She developed her own distro that would use VM to run apps from all windows environments.
I can also remember that different VM sections where separated via color and names. Such as work, home, etc.
Does anybody have a link to this?
EDIT: Thanks to Piotr_Karbowski for the answer: https://qubes-os.org/submitted by You_see_me_now
[link] [2 comments]
Reddit: One of the individuals behind the debian fork Devuan apparently released a statement in the italian forum debianizzati.org. I translated it in English for those who may be interested.
Please keep in mind that I am not in any way affiliated with the Devuan fork (actually I am an happy Arch Linux and Gentoo user). I don't have any particular opinion on systemd or Devuan, and I'm just translating because the fork was discussed here a couple of days ago.
The translation is not very smooth, in part because italian does not traslate very well to english, in part because this guy writes in a convoluted manner, in part because I don't have time to polish it.
"Hi everyone. I am Franco "nextime" Lanza, one of the individual behind the fork Devuan.
I am taking part in this thread to provide some clarifications. I hope it wil be appreciated.
I do not want to explain or discuss whether forking is the right thing to do, why we are forking, whether systemd is good or not. Upon these topics many keyboards have been worn out.
However, I would like to provide some clarifications about who we are and what we're doing, considering several concerns surfaced in this thread.
First of all, who are the "Veteran Unix Admins", or VUAs. VUAs are a group of individuals associated with a private forum, created around 3 years ago by a group of "old" sysadmins. These were all friends, mostly from the Milan area.
The name comes from: http://www.infoworld.com/article/2623488/unix/nine-traits-of-the-veteran-unix-admin.html We had an internal definition, by which a VUA is someone who identifies in at least 5 (or better, 6) of the listed points.
Notwithstanding its private nature, the VUA group has grown with time. Today, it counts 932 members as we speak; 95% are italians. The group now includes not only sysadmins, but also individuals variously involved in the IT world related to Unix.
Among these individuals, a group of around 50 people - all long time debian users, some developers and some sysadmins (some even DDs), decided to launch the (by now well known) site "debianfork.org" because, for several reasons, they absolutely did not want the imposition of systemd.
After the GR, which did not end how we wanted, we decided to proceed, and so we came to a fork. At the moment, Devuan includes among its developers around ten VUAs (the initial founders), and others have offered to join gradually as we coordinate; plus we have some developers that joined us immediately after the fork was announced.
The project is very serious, and it is not trolling. It is really nothing strange: to put it simply, we decided that the moment of whining against impositions is passed. Now is the moment to act, and we are doing it.
I want to emphasize that Devuan is NOT Debian without - or worse, "against" - systemd. Systemd will be supported in Devuan. Devuan is a fork in favor of freedom of choice. Sysvinit will remain the default init system, but all init systems will be supported - at least, all those that are packaged in Debian at the moment.
So, what changes in Devuan with respect to Debian? Many things. It has been said that systemd can be avoided in Debian, but this is not true. You can choose not to run it as PID 1, but its tentacles still remains everywhere in the system, and it's not possible to avoid it completely. We want to be able to choose to avoid it entirely.
However, I do not want to discuss the technical aspects, and I would rather answer to some of the criticisms and doubts.
First of all, it has been said that it is not clear who are the individuals behind the fork. I have explained who are the VUAs; however, there is a number of individuals who wish to remain anonymous, at least for now. This is intentional. Considering how violent is the controversy against anyone who is not "pro-systemd", many would rather not expose themselves, and for good reasons: who exposed him/herself has been subject to insults, disparagements and even threats. As for me and others, we don't care. However, for some of us this can lead to problems.
Another criticism we have received is that we have a poor online presence: guys, we are forking a huge distribution. This includes not only packages, but also infrastructure, decision-making organs, etc. The fork started a week ago, please give us enough time to organize! :D
Who will take donations: Dyne.org, a non-profit organization that has existed for many years now. It is recognized by the FSF and by the EU. It is based in Holland and operates several open source projects, including dyne:bolic. It has been known for some time for its activism in the free software world. Dyne.org has given through its Chairman Jaromil (Denis Roio, which is also a VUA member and one of the individuals behind the fork) its consent to manage the funds for us and paying taxes, all in a transparent manner (the balance sheet is public).
How we will use donations: at the moment, we do NOT need donations. We have resources donated by our members (I myself am providing 4 servers). However, it is foreseeable that in the short term we will need to ask for donations; for these kind of projects funds are always necessary. So, why not accept contributions from those willing to give them? In any case, we will publicly state how money from donations will be used.
If there are any other questions, ask away."submitted by SmellOfEmptiness
[link] [13 comments]