AJ's Techdirt Profile


About AJ

AJ's Comments comment rss

  • Feb 28, 2022 @ 07:16am

    RSS Feed encoding?

    Reading the new RSS feed via Feedly looks to have the character encoding set wrong, all non-ASCII punctuation appears as a ‘?’ character in my view of the feed.

  • Nov 10, 2021 @ 06:50pm

    Why not the FCC?

    I would have expected anyone operating a cell-site simulator to need a license from the FCC to do that, so why isn’t this firmly in their court? The cellphone companies pay lots of money for licenses to use frequencies in those wave-bands after all, so if this isn’t their jurisdiction to chase did someone write the associated regulations without this kind of thing in mind? Or is it just that their funding doesn’t allow them to actually enforce those regulations…

  • Oct 25, 2021 @ 12:57pm

    Re: Re: You got the wrong clause.

    The GPLv3 and hence also the AGPLv3 very explicitly define what they mean by “convey” (they deliberately don’t use a standard legal term, IIRC there’s an annotated version which explains why if you’re interested). Trump probably is in violation of the AGPL clause 13, but that’s not the clause cited in the article as David explained.

  • Aug 27, 2021 @ 08:59am

    Lessig’s Point

    IIRC this was Larry Lessig’s problem #1, or close to it at least. Make it so that politicians don’t have to spend much of their time fundraising for their re-elections and they might spend it on understanding the actual problems and how to solve them (maybe, if we’re lucky, once the existing job-holders are out or at least no longer feeling beholden to their buddies!).

  • Oct 22, 2020 @ 12:51pm

    Twitter’s wording could be 5% to blame

    It would be clearer that they are requiring you to read the tweet first if the notice said “You MUST read the article on Twitter before retweeting”. Using “can” implies it’s optional when it apparently isn’t.

    Not that I’m trying to excuse the idiots though...

  • Jul 21, 2020 @ 12:05pm

    Do they pay?

    Is there a legal difference between the FBI demanding the data and them paying for it? If Sabre is willing to sell this data and it is being paid for, is there anything (short of passing New US Privacy legislation) that the courts could do to stop the FBI from buying it from them anyway?

  • Nov 20, 2019 @ 04:12pm

    The wishy-washy part of the theorem:

    "to do well."

    It's a pity the last 3 words can't be made stronger somehow, although I have no suggestions how...

  • May 28, 2019 @ 11:23am

    Taste their own medicine?

    How legal/illegal is copyright trolling? If the only real cost of doing it is the court filings, why haven't opponents to the copyright maximalists ever sued politicians pushing this kind of thing for (possibly fake) copyright violations using the laws they wrote?

    This might make some interesting blog stories: Politicians indicted/convicted by the laws they wrote (or voted for at least).

  • Dec 26, 2017 @ 08:57am

    Obvious Fraud

    I just looked up my wife’s name (I had already reported the entries against mine) and I found at least 7 entries all with the “In 2015” text shown in the post and on the same 8/27 date but for different addresses across the US. There may be people with that name at those addresses but the chances of them *all* submitting the same text on the exact same date (even via some “press this button now” service) have to be effectively zero.

  • Jul 11, 2017 @ 01:12pm

    Ambiguous blog title...

    I wasn't 100% sure at first glance whether “People Would Pay A Hell Of A Lot More…” meant “People Would Have To Pay A Hell Of A Lot More…” or “People Would Be Willing To Pay A Hell Of A Lot More…”. I know Techdirt has long promulgated the latter, but in today's world ambiguity tends to feed trolls.

  • Jun 14, 2017 @ 03:17pm

    Can they be hoist by their own petard?

    If publishing links can be infringement, what's the easiest way to get a court to publish links to infringing content and hence to become infringers themselves? Presumably electronically filing something that includes an attached document full of links would get that document published on the court's website; do courts check everything included in filings?

    The problem would presumably be that whoever tries it would want to avoid any repercussions to themselves, so they should be links to material that the filer has permission to publish but no rights to pass on that permission to anyone else (such as the court). Would that give the rights-holder standing to sue the court?

    Does the law grant courts publishing rights to everything that gets filed in a legal case? (I guess there are probably rules which says a filer is responsible for redacting anything that shouldn't be made public, but IANAL so I don't know).

  • Jan 10, 2017 @ 12:34pm

    Republican action?

    So now that we have the GOP in charge of both houses and soon as POTUS, how long will it be before they take aim at CDA-230? Have the tech companies been paying^Wlobbying them enough to prevent it?

  • Aug 27, 2015 @ 06:58am

    Re: Difference

    Maybe the "King of Beers" is trying to hide a secret life as a drag queen?

  • Jul 09, 2014 @ 08:16pm

    Re: "Kilts"? Ye gods.

    The tilted things at least look like they're meant to be kilt-shaped, but the caddy's skirts are most definitely not.

  • Jun 07, 2014 @ 09:17am

    Depends on their system design

    As a foreigner I don't want to apologize for the NSA, but having worked with scientific systems that have to collect continuously arriving data I can understand why it might be technically impossible for them to stop destroying old data.

    The front end data collection process for each source is likely to be putting that raw data into a large circular buffer, such that incoming data overwrites the oldest data stored ? the length of the buffer and the average data rate thus control how long you end up keeping that raw data. While the raw data is still in that buffer it can be queried and extracted, but to stop destroying old data you would have to stop collecting any new data that is going to overwrite it.

    Now each data source that they're monitoring is going to have its own buffer like this which is probably placed very close to the point where the data is collected, and the system will be designed to do the querying and extraction locally as well. This means that the bandwidth between that buffer and the NSAs external storage (such as that big data center in Utah) can be much smaller than the incoming raw data rate, so they just can't copy all of the buffered data to offline storage; there's just too much data in the buffers for that.

    This could also explain why they claim that they're not "collecting" data on everyone; until they actually enter a query that will select a particular data item and send it back to their data center, all those circular buffers are just holding the past history temporarily and aren't doing anything with it. If they wait too long it will get overwritten, thus fulfilling their limited time legal requirements.

  • Mar 21, 2014 @ 09:13am

    It's hard to use the information though

    Anyone who gives out one of these phones has the same problem that the Allies had in WW2 with the decrypts that came out of Bletchley Park ? you have to be extremely careful how you use the information, because you won't want your target to suspect what access you have to their actions. As soon as they find out, they'll stop using the phone for the kinds of activities that you want to follow...

  • Nov 22, 2013 @ 06:34pm

    Isn't this just party politics as usual?

    If GOP were able to give the White House trade promotion authority, then after the TPP started to cause problems they could use that as political ammunition against the democrats in the future ("Obama ruined the US legal system for ever" etc.). Maybe I'm being really cynical here, but if someone doesn't care about the means (i.e. the effects the TPP would have), that end could have a political justification. From what has been leaked the USTR seems to have been listening to some of the GOP's best backers anyway, so this could be a double plus for them.

  • Mar 13, 2013 @ 08:47pm

    Re: RSS Reader

    One of the reasons I use Reader (and Bloglines before it) is because it's more efficient for a central service to poll the feeds of all its users once rather than have each user doing the same polling. I have at least 5 different machines (laptop, tablet, phone, work PCs etc.) that I use to visit Reader at different times, and I don't want to see stories more than once or have to maintain lists of feeds in multiple places, so I need cloud storage of where I've got to in each feed. I explicitly do not want my own client programs on each machine, unless they're querying a server which is following all my feeds for me.

  • Mar 13, 2013 @ 08:29pm


    Here is a CNET article that gives a list of 5 possibilities and which platforms they run on (Web, iOS, Android). I've only tried Feedly on iOS though, and as a long-time G-Reader user I wasn't too impressed, but it might work out with a bit more testing. The feedly.com website seems to be getting hammered (by all us Reader users?) just at the moment though, which could be a bad omen.

  • Dec 07, 2012 @ 01:59pm

    DMCA take-down mis-use correction?

    If this new court gets created, it should also accept cases of wrongful DMCA take-downs, and be able to force the notifier to compensate the victims. At least that could have a useful effect...

More comments from AJ >>