4 ms·

Feedly Launches Strikebreaking as a Service

50 points by jrepinc 4y ago

gruez 4y ago
I don't get it. The "Strikebreaking as a Service" described seems to be some sort of AI tool that can crawl the web and find relevant results for strikes/protests. The author thinks this is bad because it could be used for union busting purposes, but doesn't actually explain how this tool can be used for "corporate union-busting and suppression of worker organizing". I thought he whole point of striking/protesting is to get attention? I'm not sure how letting companies know that a protest/strike is happening is going to help for that, because chances are they already know.
- giraffe_lady 4y ago
  I think the author adequately and correctly focused on the framing from feedly being an important signal about how they themselves expect it to be used. The repeated implication that protests are dangerous or violent, the assumption that their customers would view protests primarily as a risk to their assets: revealing! This could have been framed as a way to find nearby protests and labor actions to participate in, for example. I point that out not to say that they should or that it would be better, but to make clear this is a choice they made based on their understanding of their customers and how they would find value in the tool, not an inherently neutral or objective or inevitable framing of the tool. I think the author is right to be unsettled by it and expressed the core concern clearly. It's not necessary to carefully enumerate every precise threat that it could play a role in. In fact it may even be counterproductive, on the internet, where being precise just exposes more surface for hobbyist contrarian nitpickers to latch on to. No technology is neutral and the expected use of this one is concerning, especially considering the admitted lack of consideration they put into it. At the very least no engineer should ever participate in a project without sitting down and thinking "what's the worst thing a malicious actor could do with this" and that clearly didn't happen here.
  - 1123581321 4y ago
    It seems more likely that Feedly is like you: understanding multiple markets exist, and want different things from the same service and those things are communicated using different words. There’s still the question of whether enterprise security budgets or protest-joining budgets are larger, admittedly.
  - gruez 4y ago
    > The repeated implication that protests are dangerous or violent, the assumption that their customers would view protests primarily as a risk to their assets: revealing! That doesn't seem like an entirely unreasonable position for their target audience (ie. enterprises) to have. Sure, not every protest is going to turn violent, but you'd still want to keep tabs on any in case they turn violent. >I think the author is right to be unsettled by it and expressed the core concern clearly. It's not necessary to carefully enumerate every precise threat that it could play a role in. In fact it may even be counterproductive, on the internet, where being precise just exposes more surface for hobbyist contrarian nitpickers to latch on to. It's fine to have a vague concern that you can't fully flesh out. But if you can't support a claim, don't pretend that you can, like by putting that into your blog post's headline.
  - bberenberg 4y ago
    I have worked for employers where there were protests outside of our building. Employees felt unsafe and in an effort to protect them, corporate security monitored this type of thing and notified them so they could choose not to come into the office. It was never about strike breaking, it was entirely about making employees feel safe.
- spondylosaurus 4y ago
  Yeah, I'm also a bit lost on the strikebreaking angle, especially if this is just pulling info from publicly accessible news sites. From the title alone (since I wasn't familiar with what Feedly is or does) I was expecting something more like an AI-powered scab—your coders or writers go on strike, some LLM tool tries to do their jobs in the meantime. Of course, knowledge workers seem to have an unfortunate aversion to striking (and unions in general), so I'm not sure how useful that kind of tool would even be. :P
j_crick 4y ago
On a bit of a trolling side, but: those who don’t have expectations can’t experience frustration. Therefore, use Inoreader. /s
jerlam 4y ago
Feedly appears to be ramping up its "AI", as I'm seeing a lot more things tagged. A lot of things also mis-tagged. I'm also seeing a lot of Feedly prompts to confirm/deny tags on super obvious items, which makes me think the technology is not very mature, or possibly it's entirely crowdsourced. "FAA issues new safety guidance amid recent ‘close calls’ at US airports" - Is this about "Travel"? "Storm-hit California slammed by another historic atmospheric river" - Is this about "Climate"? "As the housing market slowed, the Bay Area lost 4 ‘million-dollar cities’" - Is this about "Real Estate"?
acyou 4y ago
I think it is more Surveillance as a Service that Feedly is offering. You can have a problem with SaaS, but the title is wrong. Can we change it to something more accurate?
johnea 4y ago
I'm not really sure why anyone needs an "RSS Service". Can't you just use an RSS reader on your desktop or phone?
- ImPostingOnHN 4y ago
  an RSS service lets you have essentially a personalized google news that is consistent across devices (IE like google reader but not shut down)
- tetrisgm 4y ago
  It's nice to have a hosted service that will sync across multiple devices. No need to self host. Also feebly has a nice app. I use feedbin myself, but I recognize that feedly has the best ux out there for an rss reader.
deafpolygon 4y ago
I mean, if the strike is already happening - how is this a strikebreaking tool? It's not like Feedly has some unique insight and capability to predict strikes before they occur. I think the real criticism isn't so much strikebreaking but the framing of strikes as violent/dangerous. But it's not clear in the article. It's hardly a strike-breaking or a surveillance tool (since it is based on publicly reported news).
subjectsigma 4y ago
Having more data on strikes and protests isn’t a bad thing, it’s what you do with that data. The argument that Feedly is explicitly supporting strikebreaking by offering this is really flimsy. By her metric search engines and Twitter are both anti union tools. I’ve liked some of Molly’s work in the past but this seems to be mostly rage bait. She’s upset that Feedly used the wrong words and that might maybe cause something bad to happen somewhere. Cry me a river.

50 points by jrepinc 4y ago

I don't get it. The "Strikebreaking as a Service" described seems to be some sort of AI tool that can crawl the web and find relevant results for strikes/protests. The author thinks this is bad because it could be used for union busting purposes, but doesn't actually explain how this tool can be used for "corporate union-busting and suppression of worker organizing". I thought he whole point of striking/protesting is to get attention? I'm not sure how letting companies know that a protest/strike is happening is going to help for that, because chances are they already know.

giraffe_lady 4y ago
I think the author adequately and correctly focused on the framing from feedly being an important signal about how they themselves expect it to be used. The repeated implication that protests are dangerous or violent, the assumption that their customers would view protests primarily as a risk to their assets: revealing! This could have been framed as a way to find nearby protests and labor actions to participate in, for example. I point that out not to say that they should or that it would be better, but to make clear this is a choice they made based on their understanding of their customers and how they would find value in the tool, not an inherently neutral or objective or inevitable framing of the tool. I think the author is right to be unsettled by it and expressed the core concern clearly. It's not necessary to carefully enumerate every precise threat that it could play a role in. In fact it may even be counterproductive, on the internet, where being precise just exposes more surface for hobbyist contrarian nitpickers to latch on to. No technology is neutral and the expected use of this one is concerning, especially considering the admitted lack of consideration they put into it. At the very least no engineer should ever participate in a project without sitting down and thinking "what's the worst thing a malicious actor could do with this" and that clearly didn't happen here.
- 1123581321 4y ago
  It seems more likely that Feedly is like you: understanding multiple markets exist, and want different things from the same service and those things are communicated using different words. There’s still the question of whether enterprise security budgets or protest-joining budgets are larger, admittedly.
- gruez 4y ago
  > The repeated implication that protests are dangerous or violent, the assumption that their customers would view protests primarily as a risk to their assets: revealing! That doesn't seem like an entirely unreasonable position for their target audience (ie. enterprises) to have. Sure, not every protest is going to turn violent, but you'd still want to keep tabs on any in case they turn violent. >I think the author is right to be unsettled by it and expressed the core concern clearly. It's not necessary to carefully enumerate every precise threat that it could play a role in. In fact it may even be counterproductive, on the internet, where being precise just exposes more surface for hobbyist contrarian nitpickers to latch on to. It's fine to have a vague concern that you can't fully flesh out. But if you can't support a claim, don't pretend that you can, like by putting that into your blog post's headline.
- bberenberg 4y ago
  I have worked for employers where there were protests outside of our building. Employees felt unsafe and in an effort to protect them, corporate security monitored this type of thing and notified them so they could choose not to come into the office. It was never about strike breaking, it was entirely about making employees feel safe.
spondylosaurus 4y ago
Yeah, I'm also a bit lost on the strikebreaking angle, especially if this is just pulling info from publicly accessible news sites. From the title alone (since I wasn't familiar with what Feedly is or does) I was expecting something more like an AI-powered scab—your coders or writers go on strike, some LLM tool tries to do their jobs in the meantime. Of course, knowledge workers seem to have an unfortunate aversion to striking (and unions in general), so I'm not sure how useful that kind of tool would even be. :P

j_crick 4y ago

On a bit of a trolling side, but: those who don’t have expectations can’t experience frustration. Therefore, use Inoreader. /s

jerlam 4y ago

Feedly appears to be ramping up its "AI", as I'm seeing a lot more things tagged. A lot of things also mis-tagged. I'm also seeing a lot of Feedly prompts to confirm/deny tags on super obvious items, which makes me think the technology is not very mature, or possibly it's entirely crowdsourced. "FAA issues new safety guidance amid recent ‘close calls’ at US airports" - Is this about "Travel"? "Storm-hit California slammed by another historic atmospheric river" - Is this about "Climate"? "As the housing market slowed, the Bay Area lost 4 ‘million-dollar cities’" - Is this about "Real Estate"?

acyou 4y ago

I think it is more Surveillance as a Service that Feedly is offering. You can have a problem with SaaS, but the title is wrong. Can we change it to something more accurate?

johnea 4y ago

I'm not really sure why anyone needs an "RSS Service". Can't you just use an RSS reader on your desktop or phone?

ImPostingOnHN 4y ago
an RSS service lets you have essentially a personalized google news that is consistent across devices (IE like google reader but not shut down)
tetrisgm 4y ago
It's nice to have a hosted service that will sync across multiple devices. No need to self host. Also feebly has a nice app. I use feedbin myself, but I recognize that feedly has the best ux out there for an rss reader.

deafpolygon 4y ago

I mean, if the strike is already happening - how is this a strikebreaking tool? It's not like Feedly has some unique insight and capability to predict strikes before they occur. I think the real criticism isn't so much strikebreaking but the framing of strikes as violent/dangerous. But it's not clear in the article. It's hardly a strike-breaking or a surveillance tool (since it is based on publicly reported news).

subjectsigma 4y ago

Having more data on strikes and protests isn’t a bad thing, it’s what you do with that data. The argument that Feedly is explicitly supporting strikebreaking by offering this is really flimsy. By her metric search engines and Twitter are both anti union tools. I’ve liked some of Molly’s work in the past but this seems to be mostly rage bait. She’s upset that Feedly used the wrong words and that might maybe cause something bad to happen somewhere. Cry me a river.