I use a bunch of Python scripts to publish interlinked notes written with Obsidian[1] and, more recently, SiYuan[2] with Hugo. For me, this presents the best of both worlds. Obsidian and SiYuan are very nice to write and review notes. Hugo is good for publishing to the web. Having script automation to bridge the two means you can use the strengths of each tool without having to bend it for a use case it wasn’t designed for.
I love SiYuan Note.
When I discovered that exported Markdown files weren't interlinked, I wrote a Python script that uses the API and links them all together and published it on their forum.
Replace 'Security' with 'Safety' and your comment still rings eerily true! One exception (? probably not an exception) is that sometimes organizations will still hire safety/security specialists (because the externally viewed perception of not having them is damaging), but these specialists are then (perhaps unintentionally) knee-capped to various extents in order to prioritize promises to leadership or other org. goals that must be hit. The net result is that good and capable folks who care deeply about the company's mission may watch in consternation as the org. slowly train wrecks itself if the roll of the dice doesn't land right. See, for example, the story of several autonomous driving organizations.
I've seen this happen at $EMPLOYER and it actually went beyond the website. Any email you send that has the url/domain in the text (e.g. in the signature) gets flagged by gmail or any G workspace email server with a big red warning. So, all customers who use Google's email servers (directly or indirectly via G Workspace) will get the red warning banners on all emails sent from anyone in your organization. Now THAT gets annoying real quick.
> A car doesn't need data updates, and definitely not code updates
I don't think this is accurate. Many advanced driving assistance capabilities need access to updated map tiles, which is a data update. They may need code updates to fix errors or shortcomings that can be detected only after deployment on extensive fleets or in response to changes to the environment/infrastructure. This is just one example for why data and code updates are needed.
I think it is more accurate to say that a "dumb" car with mostly electro-mechanical systems doesn't need data updates and definitely not code updates. But that isn't true for vehicles built within the last few years and definitely untrue for vehicles that will be built in the coming years.
> Many advanced driving assistance capabilities need access to updated map tiles
Your phone (or GPS or even a paper map) can guide you; none of the following need access to map tiles:
* forward collision warning
* automatic emergency braking
* lane departure warning
* adaptive cruise control
* blind spot detection
* stability control
> code updates to fix errors or shortcomings
That's what recalls and TSBs have traditionally been for, and the driver can refuse them if desired. I mean, actual lives are at stake here. Would we (or should we) allow 737's to get OTA updates? Of course not. The target is too valuable and surface area too vast to adequately protect it.
The insight that worked for me is that it isn’t about the tools. Rather, it’s about having a solid plan, or methodology, that works for you, which can be implemented using whichever tools are convenient. Many folks, including me at one point, hop from one tool to another, looking for one that “feels right” and which will magically fix and organize your life.. or make you want to do that. Ain’t no such thing. But once you figure out the methodology, the rest slides into place.
Here[1] is the methodology that works for me. It enables long and short term planning and organization in accordance with my changing priorities and values.
I think it helps if personal finance managers explicitly describe at least the following:
1. What automation, if any, exists for entering transactions? This is the most laborious/cumbersome part of personal finance. Some tools use financial data aggregators (plaid, yodlee etc.) that involves sharing login credentials with a third party, sometimes disabling 2FA, or other steps that are anti-security or anti-privacy. It sucks that in the USA at least, there is practically no way for customers to fetch their bank data via an open API. Until recently, many financial institutions supported OFX, but that is being phased out.
2. How is categorization of transactions accomplished? Ideally, I want autocategorization based on my own previously categorized transactions, since the bulk of my transactions are repeats at the same merchants.
3. What sort of reporting, dashboarding, and potentially sharing capabilities exist? Ideally, I want to share some reports with my partner
A while ago, I created my own homegrown system to automate my personal finances[1]. It is capable of doing all of the above, without sharing data with a 3rd party. Unfortunately, the automated transaction retrieval mostly does not work because financial institutions are dropping support for OFX.
> 1. What automation, if any, exists for entering transactions? This is the most laborious/cumbersome part of personal finance. Some tools use financial data aggregators (plaid, yodlee etc.) that involves sharing login credentials with a third party, sometimes disabling 2FA, or other steps that are anti-security or anti-privacy. It sucks that in the USA at least, there is practically no way for customers to fetch their bank data via an open API. Until recently, many financial institutions supported OFX, but that is being phased out.
It has import support (can be templated using handlebars), no
automated fetch though
> 2. How is categorization of transactions accomplished? Ideally, I want autocategorization based on my own previously categorized transactions, since the bulk of my transactions are repeats at the same merchants.
It has a very crude tf-idf based categorization. I do have plans to improve it.
> 3. What sort of reporting, dashboarding, and potentially sharing capabilities exist? Ideally, I want to share some reports with my partner
You can checkout the https://demo.paisa.fyi It's just a web app that
works over http, so you can run it on a machine and share them
access. You have to figure out the authorization part, I do plan to
add some password based auth soon.
> It has a very crude tf-idf based categorization. I do have plans to improve it.
You could use the same algorithm that was used for spam filtering in my Linux Journal article from back in 2003[1]. I have thought about making a plugin to do it with MoneyDance but haven't gotten around to it. But I think it would be quite easy for you to integrate into Paisa if you are looking to do that sort of thing. I have actually made some improvements to the algorithm since then; let me know if you're interested...
> How is categorization of transactions accomplished? Ideally, I want autocategorization based on my own previously categorized transactions, since the bulk of my transactions are repeats at the same merchants.
One of my aggregators categorizes any alcohol purchases (purchases from a state 'ABC' store) as 'home improvement'. While technically the house looks better when I'm drunk, I still thinks it's a mistake, and I submitted feedback to them. 2 years ago. No change.
Would you rather your aggregator sell your alcoholism data to your health insurance company, or your joy of maintenance and taking good care of your property to your home insurance company?
Why not both? I would not be surprised to find the banks already sell that data to other entities anyway (directly or indirectly). I end up buying perhaps $100 of hooch per year; unsure I personally care if anyone knows, but yes, for some it might be an issue.
This "basic" guide is probably well-intentioned and useful. It does gloss over a whole range of nuances and diversity of protocols for India. Wonder if similar glossing over is happening for other countries in the list.
I think certain sounds are hard for Western speakers to pronounce, while others are easy. Specifically, Hindi (and many other Sanskrit-derived languages) distinguish between some sounds (which have correspondingly different characters in the written script) that are often indistinguishable to Western ears. And if you can't hear the difference, you are unlikely to be able to say it right.
These sounds typically come in groups of four. For example, there are 4 "T" sounds in Hindi: त थ ट ठ. If you pronounce each of these correctly, Indian ears will hear 4 distinct sounds. Western ears will typically hear the same sound 4 times. If they are listening attentively the second time around, they may be able to distinguish at most 2 different sounds, but not 4. Given this, Western speakers are unlikely to correctly pronounce a word containing one of these sounds. The good news is that even if a Western speaker has incorrect pronunciation, in the vast majority of cases native Hindi speakers would be able to understand what was meant.. so the communication still happens effectively. Furthermore, regardless of language, I like to believe that most native speakers will very much appreciate efforts to speak the language, no matter how mangled the pronunciation is.
In that table, त is the first column of the Dental row and थ in the second column of that same row (aspirated?). Similarly, ट is the first column of the Retroflex row and ठ is the second column of that row.
The alveolar ridge is a bit behind the upper front teeth, and sometimes people let blade of the tongue (tip plus a bit) rests against that when making a "t" vs against the teeth themselves.
t̪ - t with square bracket below is the dental plosive
ʰ - superscript h is the aspirations indicator
So there are three issues for English speakers:
A minor issue: English uses apical plosive /t/ for "t" , while Hindi and related use dental plosive /t̪/. This is a small difference in tongue placement. But these sound similar enough. The bigger issue is English lacks a retroflex plosive (tongue curls), and aspiration is non-phonemic (does not carry a meaningful distinction). English speakers typically aspirate leading plosives and don't aspirate in the middle of words. It's not that native English speakers "can't" hear a difference, it's just way more subtle and likely to be missed unless specifically listening for it.
This is fascinating. Thank you for commenting. As an Indian living in the West, I have always wondered why/how certain differences that are stark to my ears are subtle or barely perceptible to others.
The converse is also true. For example, the way native Swedish speakers pronounce seven (sju - example pronunciations at https://forvo.com/word/sju/ ) is 1) Hard for me to say and 2) No matter how I say it, the response from Swedes is, "You said <X>; it's actually <X1>" where both <X> and <X1> sound exactly the same to me, so I don't hear the distinction they are trying to point out. I assumed the same happens to Western folks when Indians/Hindi speakers try to explain the difference between the various T sounds.
> The bigger issue is English lacks a retroflex plosive (tongue curls), and aspiration is non-phonemic (does not carry a meaningful distinction)
But English words do seem to distinguish meaningfully between what you term 'voiceless' and 'aspirated voiceless' isn't it? For example, there is a difference between 'time' and 'thyme'. Ignoring the difference between 'y' and 'i' for a moment, wouldn't both words be the "same" to English speakers if what you are saying is true? Isn't 'th' just the aspirated version of 't'? (Not contesting what you are saying, just curious to understand.)
"Thyme" is odd because it used to be spelled "tyme" in Middle English (1066-1400s). I assume it was changed to 'th' to be more similar to Latin and French. Something similar happened to "island" (iland) and "isle" (ile) where a silent 's' was added to make the words closer to the Latin "insula".
> But English words do seem to distinguish meaningfully between what you term 'voiceless' and 'aspirated voiceless' isn't it? For example, there is a difference between 'time' and 'thyme'. Ignoring the difference between 'y' and 'i' for a moment, wouldn't both words be the "same" to English speakers if what you are saying is true?
Aspiration is not contrastive in English - it's impossible to find two words that differ only by aspiration. Aspirated consonants (in general American English) mainly feature in the onset of a stressed syllable (pin ['pʰɪn], potato [pə̥ˈtʰeɪɾoʊ]) as long as they're not preceded by /s/ (spin ['spɪn]). The important part is that you can determine whether or not a consonant is aspirated only by its position in the word, which is why it's an allophone - a variation of a phoneme which isn't distinctive, but still sounds different. English is my L2 so "thyme" still messes me up, I always try to pronounce it with /θ/ like the first consonant in "thigh".
How people differentiate sounds is actually very interesting. The leading theory is that infants can differentiate all human phonemes (see Jusczyk's Head turn Experiment) but starts categorizing sounds into categories based on what languages are spoken to them by 9-12 months. An interesting language is the (sadly extinct) Ubykh, which had 84 (!) phonemic consonants but only 2 or 3 distinct vowels. For example, speakers percieved /qʲ q qʷ qˤ qˤʷ/ as five different sounds, even though an English-only speaker would probably categorize all of them as just "kinda guttural".
On "sju" (/ɧʉː/) - /ɧ/ is a very odd sound in general. It doesn't really feature in any other languages, and what exactly it should be categorized as is still debated by phoneticians. It also varies a lot by region - Finland Swedes generally don't differentiate the consonants in "sju" and "köpa". So bottom line, we don't know how to pronounce it either :D
> For example, there is a difference between 'time' and 'thyme'. Ignoring the difference between 'y' and 'i' for a moment, wouldn't both words be the "same" to English speakers if what you are saying is true?
But they are the same; both are pronounced (in traditional IPA) as /taɪm/
There is no difference in pronunciation between time and thyme. Both will be pronounced with the same amount of aspiration. Why English maintains a th- spelling for some words of Greek and Latin origin instead of plain t- is a long story, but in English it makes no difference in pronunciation.
This is true even among western languages.
For example, it’s basically impossible for any other western european speaker to say the Portuguese ão sound.
First thing I did in my company, which is mostly German, was to abbreviate my name, instead of having multiple people trying their hardest to call me singing I don’t care for, I chose his I’m called in that context.
I rolled my own solution[1] out of existing stuff that is mostly automated, does auto-categorization of transactions, has dashboards and sharing, and does not require sharing login credentials with third parties.
However, financial institutions in the USA are making it harder to pull transactions via an API without involving third parties. So I’ll soon need to resort to scraping via selenium to automatically fetch transaction data.
I don't understand the point of traveling with clean burner devices and keeping your data encrypted in the cloud. Yes, it protects for threats where the devices are stolen or compromised when out of sight, but not for cases where government authorities are targeting you, as described in the article. What happens when govt. goons tell you to write down a list of your cloud accounts (email, storage etc.) and their corresponding security credentials and threaten you to not leave any out? Or, when you are asked to log in to your cloud accounts with the threat actors hovering around you? How many of us would refuse and/or roll the dice on not revealing certain accounts and risk them being discovered later (along with implications of not having revealed them earlier when specifically told to do so)?
Wouldn't it be more rational and reasonable (for everyday folk, not journalists, activists, dissidents, etc.) to never travel with or keep on cloud storage any data that they would rather authorities never, ever see, if at all they have such data?I think the vast variety of business and personal data does not fall into this category.
Note that, in principle, I am all for privacy and resisting govt. intrusions into private lives by crafting appropriate legal frameworks and strong technical mechanisms. In practice, as an average Joe, I don't know how much I should resist if/when I am personally targeted and threatened with dire consequences while traveling in a foreign country. It is easy to think that in such a situation, my priority would be to get out of that situation asap and folding completely may be seen as the fastest way to achieve that.
> What happens when govt. goons tell you to write down a list of your cloud accounts (email, storage etc.) and their corresponding security credentials and threaten you to not leave any out?
That's why you uninstall all apps and delete your browsing history. They have no way to know how accounts you have, or where they are. Unless you leave those traces on your devices.
If you're super paranoid about it, you can create a few cloud accounts and seed them with innocuous or otherwise fake data. That way you have something to provide, but it's nothing of interest.
> They have no way to know how accounts you have, or where they are.
Hahahahahahaha. Data brokers are happy to sell them a list of all your accounts, cell-phone location history, your credit card purchases going back 10 years, and much more.
> What happens when govt. goons tell you to write down a list of your cloud accounts (email, storage etc.) and their corresponding security credentials and threaten you to not leave any out?
Since I have no such accounts, nor social media accounts, I wonder what would happen if I were ever asked for them?
[1]: https://sagar.se/notes/computers/hugo/digital-garden/publish...
[2]: https://sagar.se/notes/computers/hugo/digital-garden/publish...