7 ms·

What exactly does a company like Evernote have to do with machine learning ? Do they have a targeted advertising platform? I can see how it might be good for th

What exactly does a company like Evernote have to do with machine learning ? Do they have a targeted advertising platform? I can see how it might be good for things like image recognition but what else are they doing with it? I am Genuinely curious, I've not used Evernote for several years, it just got out of hand and bloated I found. Machine learning is really starting to feel just like another trigger phrase.

jhayward 10y ago
Evernote's main feature when first announced was very good handwriting recognition. You could just snap a picture of a board or note pad and have searchable text in the cloud. There's some kind of cluster of OCR expertise in the Russian computer science community given ABBYY Finereader as best-in-class printed OCR and Evernote as best handwriting OCR.
- Omnipresent 10y ago
  Tesseract (used correctly with some pre-processing) can compete with ABBY
ghshephard 10y ago
Machine Learning? I'd just like to be able to do a boolean search of my notes inside the client. Something like: {Search for (notes tagged household or notes tagged personal) and (tagged 2016)} Or even something like: not notebook:"Personal Finance" I love evernote, but it often feels like the team writing the search engine quit before the job was quite done.
- mschnell 10y ago
  Try -notebook:...
  - ghshephard 10y ago
    Nope, no such luck. You can search for things that are in a notebook with the query: notebook:receipt but not -notebook:receipt
Sylos 10y ago
Yeah, Evernote being bloated was also the reason for me to stop using it. I wanted to take a few notes here and there, not document my entire life in all available forms of media. I can't believe that they are still adding even more things onto it.
Infernal 10y ago
Their paid features include OCR, both for handwriting and images of printed text. Seems like a textbook machine learning problem. https://blog.evernote.com/blog/2015/01/23/search-handwriting-evernote/ https://blog.evernote.com/blog/2015/01/23/search-handwriting...
- bdavisx 10y ago
  And I've been very, very impressed with the handwriting recognition. I discovered it by accident before it was a paid feature when I was searching for something and it brought up an image of some scribbled notes I had taken with the search highlighted. And my handwriting is bad, really bad.
  - afro88 10y ago
    OCR with Evernote works very well because it's allowed to be fuzzy. One hand written word is allowed to have multiple predicted texts. This is because the predictions are only used for search - you can't copy/paste OCR'd text because Evernote can't guarantee it's correct. But it's good enough for fuzzy search. In fact it's pretty much exactly what you want for search: allowances for misspellings and typos. Very smart.
  - danso 10y ago
    Awhile ago someone asked how a product like ABBYY Finereader, which does great when OCRing typed text, does on handwriting. Not well at all I'm afraid: https://gist.github.com/dannguyen/50dcb5a8f4230e7a8a40bfe2d0a1c836 https://gist.github.com/dannguyen/50dcb5a8f4230e7a8a40bfe2d0...
    - praseodym 10y ago
      Evernote doesn't just OCR handwriting to a single text; it indexes a bunch of possibilities for every handwritten word. This way the search functionality can search through all these variations, not just the most likely one.
      - 10y ago
sidlls 10y ago
>Machine learning is really starting to feel just like another trigger phrase. Yes, and its close cousin "data science" fits the same description, to me. They're quickly becoming what "the cloud" is: marketing jargon that doesn't provide useful information.
- derefr 10y ago
  Both of those terms are indeed "marketing jargon", but I don't think they're intended to market products—rather, they market people. The point of treating data science as a field is to distinguish the people doing it from "regular programmers." This creates a distinction of caste: the "regular programmers" are relegated to maintenance work, never trusted to touch any of the mathematical and CS-degree-requiring parts of their own discipline. The "data scientists" are then hired from outside, to do precisely the parts of programming that every programmer at the company would have already been proved competent at, through degree-qualification and technical interviews that seek out exactly these talents. And the "data scientists" are paid better for it, and given better working conditions, and less micromanagement. Same with machine learning: it's just programming—certainly, a kind of programming that is heavy on CS, but certainly one you can pick up quickly enough to get some good practical results. But you hire ML people separately from "programmers", and you never hand your "programmers" ML tasks. From yesterday's subthread on the wind-down of Starfighter, I gather that a large part of its goal was to reveal that the emperor has no clothes here: that nearly any programmer is capable of doing "data science" or "machine learning" (and, as well, "reverse-engineering", "low-level programming", and a host of other things) if actually given a chance to solve the problems.
  - fixermark 10y ago
    The logic here seems flawed; it's generally cheaper and faster to hire from within than without (you don't have to spin up people on corporate process or software that someone else wrote). So why wouldn't companies do that? And the flaw in the logic is revealed here, I think: > nearly any programmer is capable of doing "data science" or "machine learning" Nearly any programmer might very well be capable of learning these disciplines, but they're different disciplines from business logic or database theory or networking; they're non-discrete and statistics-interpretation heavy, and it's entirely plausible that you can have an entire company full of programmers who are neither excited about nor prepared to take on these categories of challenge. Some programmers will (lots of people are polymaths), but it's not nonsense to consider them different categories of discipline. I think smart companies provide an internal channel to get into these sub-disciplines, but it's not necessarily a problem space you can throw at any developer and say "Here you go; good luck tuning those training and test sets!"
awinograd 10y ago
Possible applications that come to mind are image recognition, note classification (e.g. recipe, web tutorial, news), text search, and text recognition in things like PDFs
gumby 10y ago
They have some context searching features -- so when you open or type a note it displays (off to the side) other notes, wsj articles and the like that are supposed to be contextually related. I suppressed this because I found it a distraction, but if their searching and matching gets better I might un-disable it and give it another try.
srikz 10y ago
From the article: Evernote’s new CTO Anirban Kundu told me the first two areas that will be replaced by Google’s machine learning APIs are its voice recognition for speech-to-text translations; and natural language processing, used to help search for contextual content. I think there is a good opportunity in helping users with notes categorization using ML.
- hollander 10y ago
  Yeah, now the readable multiline version: > Evernote’s new CTO Anirban Kundu told me the first two areas that will be replaced by Google’s machine learning APIs are its voice recognition for speech-to-text translations; and natural language processing, used to help search for contextual content.
  - wincy 10y ago
    Thanks. On mobile these single line quotes are really frustrating. Is this just considered to be fine in the design of HN? Sort of a meta post sorry if it's not appropriate here.
  - srikz 10y ago
    I'm sorry that it was unreadable. I thought the double space thingy was the way to quote things here. I guess it's for code.

jhayward 10y ago

Evernote's main feature when first announced was very good handwriting recognition. You could just snap a picture of a board or note pad and have searchable text in the cloud. There's some kind of cluster of OCR expertise in the Russian computer science community given ABBYY Finereader as best-in-class printed OCR and Evernote as best handwriting OCR.

Omnipresent 10y ago
Tesseract (used correctly with some pre-processing) can compete with ABBY

ghshephard 10y ago

Machine Learning? I'd just like to be able to do a boolean search of my notes inside the client. Something like: {Search for (notes tagged household or notes tagged personal) and (tagged 2016)} Or even something like: not notebook:"Personal Finance" I love evernote, but it often feels like the team writing the search engine quit before the job was quite done.

mschnell 10y ago
Try -notebook:...
- ghshephard 10y ago
  Nope, no such luck. You can search for things that are in a notebook with the query: notebook:receipt but not -notebook:receipt

Sylos 10y ago

Yeah, Evernote being bloated was also the reason for me to stop using it. I wanted to take a few notes here and there, not document my entire life in all available forms of media. I can't believe that they are still adding even more things onto it.

Infernal 10y ago

Their paid features include OCR, both for handwriting and images of printed text. Seems like a textbook machine learning problem. https://blog.evernote.com/blog/2015/01/23/search-handwriting-evernote/ https://blog.evernote.com/blog/2015/01/23/search-handwriting...

bdavisx 10y ago
And I've been very, very impressed with the handwriting recognition. I discovered it by accident before it was a paid feature when I was searching for something and it brought up an image of some scribbled notes I had taken with the search highlighted. And my handwriting is bad, really bad.
- afro88 10y ago
  OCR with Evernote works very well because it's allowed to be fuzzy. One hand written word is allowed to have multiple predicted texts. This is because the predictions are only used for search - you can't copy/paste OCR'd text because Evernote can't guarantee it's correct. But it's good enough for fuzzy search. In fact it's pretty much exactly what you want for search: allowances for misspellings and typos. Very smart.
- danso 10y ago
  Awhile ago someone asked how a product like ABBYY Finereader, which does great when OCRing typed text, does on handwriting. Not well at all I'm afraid: https://gist.github.com/dannguyen/50dcb5a8f4230e7a8a40bfe2d0a1c836 https://gist.github.com/dannguyen/50dcb5a8f4230e7a8a40bfe2d0...
  - praseodym 10y ago
    Evernote doesn't just OCR handwriting to a single text; it indexes a bunch of possibilities for every handwritten word. This way the search functionality can search through all these variations, not just the most likely one.
    - 10y ago

sidlls 10y ago

>Machine learning is really starting to feel just like another trigger phrase. Yes, and its close cousin "data science" fits the same description, to me. They're quickly becoming what "the cloud" is: marketing jargon that doesn't provide useful information.

derefr 10y ago
Both of those terms are indeed "marketing jargon", but I don't think they're intended to market products—rather, they market people. The point of treating data science as a field is to distinguish the people doing it from "regular programmers." This creates a distinction of caste: the "regular programmers" are relegated to maintenance work, never trusted to touch any of the mathematical and CS-degree-requiring parts of their own discipline. The "data scientists" are then hired from outside, to do precisely the parts of programming that every programmer at the company would have already been proved competent at, through degree-qualification and technical interviews that seek out exactly these talents. And the "data scientists" are paid better for it, and given better working conditions, and less micromanagement. Same with machine learning: it's just programming—certainly, a kind of programming that is heavy on CS, but certainly one you can pick up quickly enough to get some good practical results. But you hire ML people separately from "programmers", and you never hand your "programmers" ML tasks. From yesterday's subthread on the wind-down of Starfighter, I gather that a large part of its goal was to reveal that the emperor has no clothes here: that nearly any programmer is capable of doing "data science" or "machine learning" (and, as well, "reverse-engineering", "low-level programming", and a host of other things) if actually given a chance to solve the problems.
- fixermark 10y ago
  The logic here seems flawed; it's generally cheaper and faster to hire from within than without (you don't have to spin up people on corporate process or software that someone else wrote). So why wouldn't companies do that? And the flaw in the logic is revealed here, I think: > nearly any programmer is capable of doing "data science" or "machine learning" Nearly any programmer might very well be capable of learning these disciplines, but they're different disciplines from business logic or database theory or networking; they're non-discrete and statistics-interpretation heavy, and it's entirely plausible that you can have an entire company full of programmers who are neither excited about nor prepared to take on these categories of challenge. Some programmers will (lots of people are polymaths), but it's not nonsense to consider them different categories of discipline. I think smart companies provide an internal channel to get into these sub-disciplines, but it's not necessarily a problem space you can throw at any developer and say "Here you go; good luck tuning those training and test sets!"

awinograd 10y ago

Possible applications that come to mind are image recognition, note classification (e.g. recipe, web tutorial, news), text search, and text recognition in things like PDFs

gumby 10y ago

They have some context searching features -- so when you open or type a note it displays (off to the side) other notes, wsj articles and the like that are supposed to be contextually related. I suppressed this because I found it a distraction, but if their searching and matching gets better I might un-disable it and give it another try.

srikz 10y ago

From the article: Evernote’s new CTO Anirban Kundu told me the first two areas that will be replaced by Google’s machine learning APIs are its voice recognition for speech-to-text translations; and natural language processing, used to help search for contextual content. I think there is a good opportunity in helping users with notes categorization using ML.

hollander 10y ago
Yeah, now the readable multiline version: > Evernote’s new CTO Anirban Kundu told me the first two areas that will be replaced by Google’s machine learning APIs are its voice recognition for speech-to-text translations; and natural language processing, used to help search for contextual content.
- wincy 10y ago
  Thanks. On mobile these single line quotes are really frustrating. Is this just considered to be fine in the design of HN? Sort of a meta post sorry if it's not appropriate here.
- srikz 10y ago
  I'm sorry that it was unreadable. I thought the double space thingy was the way to quote things here. I guess it's for code.