HNHacker News
TopNewBestAskShowJobs

haukem

197 karma · joined January 2, 2021

submissionscomments
haukem··on “Open source” is not broken
The legal and fiance constructs of many open source projects are pretty badly organized. This makes it hard for accounts at big cooperation to handle them. Many open source projects are also pretty bad at marketing to non developers and do not have an aggressive founding campaign.

Some “social” organizations are much better at this and have very good people taking care of marketing and accounting to people who control money.

Most members of open source projects just want to code and not take care of tax exception for donations and filing out the correct tax forms in time.

haukem··on AWS Graviton 3 Instances
Thanks for the document, so the ARM marketing just confused me. ;-)

Yes N2 is more likely than V1. N2 has the better PPA ratio. Own CPU core is very unlikely as I am not aware of any rumors which we would have notice before.

N2 also supports ARMv9 which is nice.

haukem··on AWS Graviton 3 Instances
Does Graviton3 use the Neoverse V1? The Graviton2 used the Neoverse N1.

The features listed here match the core: https://developer.arm.com/ip-products/processors/neoverse/ne...

The N2 misses the bfloat, but it could be that the ARM marketing named it differently: https://developer.arm.com/ip-products/processors/neoverse/ne...

haukem··on The falsehoods of anti-AGPL propaganda (2020)
From my experience most commercial software licenses are much harder to understand. Especially if you want to redistribute commercial software and do not want to pay for it, it gets very complicated.
haukem··on The falsehoods of anti-AGPL propaganda (2020)
I think it is the same as the GPL v3, at least these sections do not differ. See section 6.

If you do not want to provide the user directly a current copy of your source code to downloaded, you can "accompanied [it] by a written offer, valid for at least three years and valid for as long as you offer spare parts or customer support for that product model, to give anyone who possesses the object code either (1) a copy of the Corresponding Source for all the software".

Then you should be able to fulfill a request for the source code used at an exact date and time 2 years ago even when you already discontinued the product and let go of all people involved. If such a request comes from controlled country like Iran or a recently banned company like Huawei or ZTE it could also get funny for your legal department.

I am not a lawyer, please consult one if you want to find loopholes in contracts.

haukem··on The falsehoods of anti-AGPL propaganda (2020)
I think this answer is wrong or very dangerous, but I am not a lawyer.

The AGPL v3 says:

>... your modified version must prominently offer all users interacting with it remotely through a computer network .... an opportunity to receive the Corresponding Source of your version by providing access to the Corresponding Source from a network server at no charge.....

> if there is a security issue and it has an emergency patch but there is some coordinated disclosure going on so its not public yet, is this an agpl violation?

When you have the permission from all copyright holder to make an exception to the AGPL you do not have to publish the source code. Otherwise you have to publish it. If it is an open source project with many contributors and the copyright stays at the original authors getting such a permission is practically impossible.

haukem··on The falsehoods of anti-AGPL propaganda (2020)
The main difference between the GPL v3 and the AGPL v3 is this addition:

> 13. Remote Network Interaction; Use with the GNU General Public License. Notwithstanding any other provision of this License, if you modify the Program, your modified version must prominently offer all users interacting with it remotely through a computer network (if your version supports such interaction) an opportunity to receive the Corresponding Source of your version by providing access to the Corresponding Source from a network server at no charge, through some standard or customary means of facilitating copying of software. This Corresponding Source shall include the Corresponding Source for any work covered by version 3 of the GNU General Public License that is incorporated pursuant to the following paragraph.

Here is the full diff between the two license texts: https://www.diffchecker.com/cV5JfP5H

haukem··on Google Pixel 6 and Pixel 6 Pro
Intel does this all the time, but they want normal Ubuntu, RedHat and so on to support their hardware when their hardware gets into the market. With these phones the hardware and the software is shipped together in a bundle, it is a different thing.
haukem··on Google Pixel 6 and Pixel 6 Pro
I haven't see this as a problem, but I am working for a consumer network equipment semiconductor company. I haven't worked with graphics drivers which could be different. Normally we get the driver code from Cadence and Synopsys under a permissive license, it can be integrated in what ever you want. The documentation and especially the RTL is under strict NDA.

This driver code is often very self contained and does not use many or any Linux frameworks, it should be easy to integrate it into any operating system in any way. Normally you have to rewrite the driver code you get from Cadence and Synopsys to get it integrated in upstream Linux, because it does not meet the upstream Linux guidelines. This is a general problem with the out of tree drivers you get from the semiconductor industry.

There are also big players in the semiconductor industry which demand that every code inside the Linux kernel they ship has to be under GPL for legal compliance.

There is also not a single bad guy in the semiconductor industry which prevents upstream Linux support. Every player could do it, Google probably got most of the drivers in source for their Pixel phones and could have upstreamed them, but most of them probably need a rewrite. They could have offered Qualcomm some money to port support for the SoC used in a Pixel phone to a more recent major kernel version, I am pretty sure Qualcomm would have done it for the right amount of money.

haukem··on Juniper breach mystery starts to clear with new details on hackers and U.S. role
Wikipedia already said that Dual EC DRBG is shitty in November 2007: https://en.wikipedia.org/w/index.php?title=Dual_EC_DRBG&oldi... (article from 16. November 2007)

Whoever integrated this into their products after November 2007 is either incompetent or was bribed by the NSA.

haukem··on Big company tale: six months for a list and a button
Yes I would say this is a case when someone wants to contribute something to an open source project.

Often these initial questions are very unspecific and often as a maintainer you only get the question, answer it and then never see this person again. Time wasted.

When someone already wrote code and want to contribute it you know that the person already invested some time and is seriously about implementing some stuff. Here the likelihood that your time is wasted is much less.

haukem··on How to bypass Sprint/T-Mobile 2FA in under 5 minutes
Why are only very few services supporting WebAuthn?

Google and github supprot WebAuthn, but most services only support OATH-TOTP, OATH-HOTP or even worse only their own app.

haukem··on Security researcher found Wi-Fi vulnerabilities that existed since the beginning
More usage of HTTPS helps Google's business.

1.) ISPs can not change the content of websites their users are watching to modify or add additional advertisements. Companies have to buy advertisements at Google.

2.) It is harder for ISP to analyze the web traffic of their customers to build profiles which they can sell. Only Google has these information.

3.) When people feel safer in the Internet they buy more stuff on the Internet. Business will buy more advertisements on the Internet, probably at Google.

haukem··on Microsoft is shutting down its Azure Blockchain Service
They prove that the hash they publish in the NYTimes exists at that time. There are probably some archives which can give you an authorized copy of the NYTimes from a specific date. You can use that to prove that the hash existed at the date the NYTimes was published. This is now the hash over a hash tree, which this companies stores. With this single hash they can also prove that every document in this hash tree existed at that time. They can not add a extra document later, because this would alter the hash they published in the NYTimes.

When you publish your phony ad in the NYTimes, you also prove that this phony ad existed at the date the NYTimes was published.

This company probably buys one edition of the NYTimes with their add and check if the NYTimes published the correct hash, if it is wrong they will run an other add the next day with the correct hash the next day. Now they only prove that the hash existed the next day.

haukem··on Microsoft is shutting down its Azure Blockchain Service
You can also set up a git repository and prevent history rewrites on the central server. Some third party could also offer this as a service. Each entity can check this by doing a pull every minute in a cron job, which fails when history rewrite happened. If they detect this, their legal department can take some actions.
haukem··on Microsoft is shutting down its Azure Blockchain Service
What is the advantage of a private blockchain over a hash tree like used in git? A git commit is also irrevocable, except if you do a git rebase. How do you prevent something similar like a git rebase in a private blockchain?

A public blockchain with multiple agents that do not trust each other like in BitCoin make sense to me.

haukem··on FragAttacks: new security vulnerabilities that affect wi-fi devices
I think such long embargoes are bad.

Embargoes prevent that the average cyber criminal knows about the problems, but the resourceful organizations already get the information before the public knows about them. I think even 90 days are pretty long.

For example 253 vendors were informed about the problem in dnsmasq about 3 months before it was published: https://www.kb.cert.org/vuls/id/434904 (all vendors listed here were informed) In each organization probably multiple people know about this.

haukem··on FragAttacks: new security vulnerabilities that affect wi-fi devices
Often many companies and organizations from many countries are getting informed about such security problems under the embargo. I assume that the intelligence agencies form many countries are also getting these information. Either they are officially informed, because they also protect the government networks or they have just good working relationships with their local companies.

I assume Microsoft would inform the NSA about such things, Huawei would inform the Chinese intelligence agencies and Siemens would inform the German BND.

haukem··on Huawei Tops Intel as Top Linux Kernel Contributor for 5.10
Now Huawei is fixing bugs in SELinux: https://git.kernel.org/linus/68dae71b7cde628a082e982c0abbf6e...

Or reports them, so the NSA can fix them: https://git.kernel.org/linus/3a2f5a59a695a73e0cde9a61e0feae5...

haukem··on Huawei Tops Intel as Top Linux Kernel Contributor for 5.10
Normally the identity of contributors is not really checked when you want to contribute to an open source project. A valid email address and a name is sufficient to contribute to the Linux kernel.

All changes to The Linux kernel are getting publicly reviewed, so the hard thing is passing this technical review if you want to introduce a backdoor. If people know you, based on previous contributions, the review could be less hard, so it could make sense to build up a reputation before. In the Linux kernel not the company builds up the reputation, but the individual engineer who contributes. If this is a Huawei hardware it could be easier for an engineer with a Huawei mail address to get changes integrated, because people assume he known the driver and hardware well and people trust claims like, "this can not happen because the closed source firmware takes care of this".

If the Chinese intelligence agency wants to integrate a backdoor into mainline Linux it would be easier to just build up some own fake identities and use them. It probably would be easier to just analyze the Linux kernel and use bugs introduced by other people.

The NSA also contributed to the Linux kernel, look for contributes from a nsa.gov mail address.

haukem··on CVE Stuffing
How to mark a CVE as invalid or request an update? I tried the Update Published CVE process, but nothing happened not even a reject, just no answer. Multiple CVEs where reported to OpenWrt which are invalid, but we (OpenWrt team) haven't found out how to inform Mitre.

For example CVE-2018-11116: Someone configures an ACL to allow everything and then code executing is possible like expected: https://forum.openwrt.org/t/rpcd-vulnerability-reported-on-v...

and CVE-2019-15513: The bug was fixed in OpenWrt 15.05.1 in 2015: https://lists.openwrt.org/pipermail/openwrt-devel/2019-Novem...

For both CVEs we were not informed, the first one someone asked in the OpenWrt forum about the details of this CVE and we were not even aware that there is one. The second one I saw in a public presentation from a security company mentioning 4 CVEs on OpenWrt and I was only aware of 3.

When we or a real security researcher request a CVE for a real problem as an organization it often takes weeks till we get it, we released some security updates without a CVE, because we didn't want to wait so long. It would also be nice to update them later to contain a link to our detailed security report.

← PreviousPage 2 of 2