6 ms·

Watching this recording I really wished they would invest more in audio equipment and in people who know how to record it. Reminds me of a lecturer at uni who h

Watching this recording I really wished they would invest more in audio equipment and in people who know how to record it. Reminds me of a lecturer at uni who had a bunch of recorded quest lectures that he would show every year. Audio on those was terrible, constantly clipped sound, but when we complained he said he couldn't hear anything wrong.

MHordecki 9y ago
This is why text articles are superior to videos - or at least an accompanying transcript. Text is very hard to mess up.
- chrismorgan 9y ago
  U+1F928 FACE WITH ONE EYEBROW RAISED (Interpretation of that raised eyebrow: text isn’t hard to mess up. It is nonetheless more accessible in general, I just couldn’t leave the “text is very hard to mess up” unremarked on.)
  - scrumper 9y ago
    Your counterpoint example to the proposition that text is hard to mess up is... a picture? Edit: or have I completely missed the point of what you're saying!
    - kbenson 9y ago
      It's a picture as much as any particular character in a font is, which is to say, it is. We're just very used to recognizing particular variations and combinations. There's one problem here that can be expressed in two ways, one is a proliferation of many new characters which people may not know the idiomatic usage of until they've encountered it in more context, the other is new combinations of known characters and words into a new idiomatic usage that has the same problem. Consider that there was a time not so long ago where the vast majority of people didn't know what LOL meant and would be confused when encountering it. On the other end you could probably copy some heavy British slang in here right now and I would be very confused.
    - chrismorgan 9y ago
      My response was simply a raised eyebrow. It wasn’t a counterpoint. The fact that it was open to several additional interpretations merely suited my sense of humour.
- IshKebab 9y ago
  I would disagree for research - text is usually in the form of papers. When you're writing a paper there's a strong incentive to make sound it as complicated as possible so very few papers are easy to follow. Giving a presentation though there isn't such an incentive and the goal is more explicitly to explain your idea rather than to prove how clever it is. So I find presentations easier to follow in general (unless they are a bad speaker, of which there are unfortunately many).
jldugger 9y ago
Honestly, I would be happy to see a podcast / Youtube series come out that finds quality tech talk speakers, and polishes them up to TED level: concise 20 minute videos, with top notch audio, well rehearsed presenters and informative content.
- enturn 9y ago
  Have a look at InfoQ website. I've watched lots of great tech presentations on there. Also Gary Bernhardt from Destroy All Software.
  - jldugger 9y ago
    InfoQ looks like they just recorded your typical conference presentation. Most of the videos I'm seeing are 2x the length I'm talking about, and are half vendor pitch. Destroy all software is closer, but is kind of like one guy. Great for him! And for me on my flight here in a few minutes. But what I envision is more like training speakers to be concise and engaging yet informative. Kinda like how an editor helps a writer cut out things to make a better novel, but for tech talks. A tall order, I know.
agumonkey 9y ago
That always makes me crazy; how a "simple" thing like sounds makes a talk useful or unbearable. I always fantasize about a collaborative app that would blend a few smartphone audio recordings to cancel noise and room reverb into a nice vocal signal.
- steamer25 9y ago
  The app seems pretty feasible. There'd probably be some issues with having to periodically resync the timing due to moving microphones or other skews coming in and out of phase but I don't imagine that'd be insurmountable. The bigger trick would be getting various parties at any given event to help out by first recording and then submitting their data to the same collaboration service. I've seen enough rooms full of techies struggling to get a projector or conference call working to imagine all kinds of headaches with connectivity, app installation, finding the right event reference, etc. If you allow post-event submission, most people would probably forget or not care by the time they got back to their home networks only to be annoyed weeks or months later that they still have a large audio file taking up space on their device. As more people are posting various captures (just for personal-reasons/vanity, without necessarily going out of their way to prescribe any particular remix), maybe something like PhotoSynth where you crawl the web for media having similar GPS and timestamps could work. Maybe people could be bothered to tag their posts with hashtag #YetAnotherCon2017; then you just have to monitor a couple of the more popular social channels.
  - agumonkey 9y ago
    With accelerometer you may have enough to recalibrate. About cooperation, I'm sure it's not hard to find 10 people willing to contribute. You can even count the people talking or organizers.
- IshKebab 9y ago
  Pretty sure that already exists. Maybe not in a production app but I bet there are lots of PoCs. Here's one for example: https://twei7.com/papers/Sur_Wei_MobiSys14_Dia.pdf https://twei7.com/papers/Sur_Wei_MobiSys14_Dia.pdf

MHordecki 9y ago

This is why text articles are superior to videos - or at least an accompanying transcript. Text is very hard to mess up.

chrismorgan 9y ago
U+1F928 FACE WITH ONE EYEBROW RAISED (Interpretation of that raised eyebrow: text isn’t hard to mess up. It is nonetheless more accessible in general, I just couldn’t leave the “text is very hard to mess up” unremarked on.)
- scrumper 9y ago
  Your counterpoint example to the proposition that text is hard to mess up is... a picture? Edit: or have I completely missed the point of what you're saying!
  - kbenson 9y ago
    It's a picture as much as any particular character in a font is, which is to say, it is. We're just very used to recognizing particular variations and combinations. There's one problem here that can be expressed in two ways, one is a proliferation of many new characters which people may not know the idiomatic usage of until they've encountered it in more context, the other is new combinations of known characters and words into a new idiomatic usage that has the same problem. Consider that there was a time not so long ago where the vast majority of people didn't know what LOL meant and would be confused when encountering it. On the other end you could probably copy some heavy British slang in here right now and I would be very confused.
  - chrismorgan 9y ago
    My response was simply a raised eyebrow. It wasn’t a counterpoint. The fact that it was open to several additional interpretations merely suited my sense of humour.
IshKebab 9y ago
I would disagree for research - text is usually in the form of papers. When you're writing a paper there's a strong incentive to make sound it as complicated as possible so very few papers are easy to follow. Giving a presentation though there isn't such an incentive and the goal is more explicitly to explain your idea rather than to prove how clever it is. So I find presentations easier to follow in general (unless they are a bad speaker, of which there are unfortunately many).

jldugger 9y ago

Honestly, I would be happy to see a podcast / Youtube series come out that finds quality tech talk speakers, and polishes them up to TED level: concise 20 minute videos, with top notch audio, well rehearsed presenters and informative content.

enturn 9y ago
Have a look at InfoQ website. I've watched lots of great tech presentations on there. Also Gary Bernhardt from Destroy All Software.
- jldugger 9y ago
  InfoQ looks like they just recorded your typical conference presentation. Most of the videos I'm seeing are 2x the length I'm talking about, and are half vendor pitch. Destroy all software is closer, but is kind of like one guy. Great for him! And for me on my flight here in a few minutes. But what I envision is more like training speakers to be concise and engaging yet informative. Kinda like how an editor helps a writer cut out things to make a better novel, but for tech talks. A tall order, I know.

agumonkey 9y ago

That always makes me crazy; how a "simple" thing like sounds makes a talk useful or unbearable. I always fantasize about a collaborative app that would blend a few smartphone audio recordings to cancel noise and room reverb into a nice vocal signal.

steamer25 9y ago
The app seems pretty feasible. There'd probably be some issues with having to periodically resync the timing due to moving microphones or other skews coming in and out of phase but I don't imagine that'd be insurmountable. The bigger trick would be getting various parties at any given event to help out by first recording and then submitting their data to the same collaboration service. I've seen enough rooms full of techies struggling to get a projector or conference call working to imagine all kinds of headaches with connectivity, app installation, finding the right event reference, etc. If you allow post-event submission, most people would probably forget or not care by the time they got back to their home networks only to be annoyed weeks or months later that they still have a large audio file taking up space on their device. As more people are posting various captures (just for personal-reasons/vanity, without necessarily going out of their way to prescribe any particular remix), maybe something like PhotoSynth where you crawl the web for media having similar GPS and timestamps could work. Maybe people could be bothered to tag their posts with hashtag #YetAnotherCon2017; then you just have to monitor a couple of the more popular social channels.
- agumonkey 9y ago
  With accelerometer you may have enough to recalibrate. About cooperation, I'm sure it's not hard to find 10 people willing to contribute. You can even count the people talking or organizers.
IshKebab 9y ago
Pretty sure that already exists. Maybe not in a production app but I bet there are lots of PoCs. Here's one for example: https://twei7.com/papers/Sur_Wei_MobiSys14_Dia.pdf https://twei7.com/papers/Sur_Wei_MobiSys14_Dia.pdf