I've to say - videos in that format are very very useful to me. Enough information to form an opinion if a software is worth the time investing to learn more or even consider using it in a new tech stack. But the information is not too deep to get lost.Thank you, a very good video.
Seconded! While the 100 seconds videos are always a joy to watch, I _sometimes_ get an impression that's quite different from the actual experience with the actual tool. This is like the 101 that gives a look and feel of the thing in practice.
Even if the performance gains don't arrive, it's still great. Not every project needs to handle millions of records. For projects that handle several thousand records the performance will be totally acceptable and the relationship syntax will greatly reduce development complexity. This seems like a must have in my toolbelt.
Definitely agree. Being able to write intuitive and easy-to-maintain code is also very important, especially if you only have a very limited number of concurrently active users.
Not sure I would agree here. It's a database - not scaling is not an option. I would store my data somewhere else if fast & synchronised access is not needed.
True, I most likely don't need the best performing stuff, just need something to get it working , instead of spending precious time developing more complex and hard to maintain tech stacks.
I suppose if you're sure you'll never need to scale it, but most projects worth working on do have some scaling goals at some point. Not sure I'd want to invest in this technology then have to re-write it all in the future.
For anyone having problems following this quick tutorial (with Postman)....there are certain specific headers you should add (alongside basic auth) In particular: NS - {{namespace}} DB - {{database}} Accept - application/json The Accept-app/json header is necessary with the beta-version...for some reason....but adding it will make it finally work
Man, I'm barely a developer and Surreal looks so goddamned awesome. I'm starting to think about a career change from architecture and if their promises hold up, I could definitely see myself getting deeper into this.
Same dude. Except I'm not an architect - I mainly sit around in my underpants watching RUclips videos, but this makes me consider giving all that up...
Think twice before jumping in development. Companies are laying off and the next decade might not be as bright as the last one. Cheers if you guys still make the switch :)
My man talks about changing his life around. I mean rn he sits in his underpants doin nothing and he thinks about developing, you tell him to think twice??? What did companies/colleague did to you man WTF?
@@ThunderboltPath This has always happened. Recruiting and laying off is just a cycle when it comes to IT and does not mean that things will get harder for future developers. This is like clamining that "AI will steal our jobs". It just won't.
@JT wanna swap? Been a software engineer for 10+ years and writing code for 20. Wish I’d studied architecture and as building tangible things in the real world. Agree, SurrealDB looks Sick! 😍
The DB does look amazing, the queries seem so intuitive to write - I guess the only danger of it is that how easy it makes adding data that one might just no consider a proper structure before writing the code... But man does it look like magic...
It's also like saying that exposed wiring makes it so easy to do electrical repairs, but that one might accidentally shock themselves if they aren't careful. Making it hard to foot-gun yourself isn't a bad thing in itself. It's at which cost we should argue about. Some will prefer the freedom of schemaless. Others will prefer the structured nature of a schema. I lean towards the latter.
@@pjarnfelt in your analogy choosing to use a fork instead of a spoon suggests it will take longer to program with a schema based db , however being burned represents your query’s being so in performant many fail and users are left waiting multiple seconds for anything to load on top of their own internets speed, and or crashes start happening from lack of ram to support the quantity of requests for how long each ones taking. And although as the developer you might not care about a little burn, programming is more like being a chef. Are you gonna serve your soup while it’s still boiling, or make sure it’s just hot? Cause you won’t have many users returning if their experience is being burned by a painstaking slow website or application because of non thought through database scheme or just lack of one
@@fluffio2976 If your application could kill someone (i.e. you're in critical systems) then you should be using an enforced schema and other safety checks. 99% of our applications are _not_ critical systems, yet our DB tools are designed as if they are. How many decades did it take for someone to design something intuitive like this?
My brother in law told me about SurrealDB last week. Having worked with ArangoDB before, its extremely similar in my opinion but with a future that looks very bright. Really liking what I'm seeing from it so far!
I love ArangoDB which is much further into development (mature) but this looks more intuitive! I'll definitely be following this project...it has that new, shiny vibe! LOL
It has basically everything you;d want which is amazing. I cannot believe how harder it will be to succeed though focusing on so many features. Will definitely use it in a future side project
Modern DBs are what they are for a reason. Have fun diagnosing your performance issues on a production database after a dozen of developers have crafted the most obtuse and convoluted queries possible. The syntax might be simple on the surface but the generated queries under the hood are going to explode quickly
I’d be really curious about performance. If it can match a Postgres, I don’t see a reason to ever use traditional relational databases again. However I doubt it can if you really want to take advantage of those traversal methods.
It seems like it's built where you can pick the tradeoffs you are willing to live with. Some want "all teh speed" while others just want enough speed to be reasonable, but relatable code.
Man! A "deep dive" and it's only a little less than 8mins. You do amazing content and I know quality takes time but please moooore of this! I love this new channel idea btw!
I love this format. You are keeping me relevant. Your videos are a great educational tool for continued learning. I'm building an app in tauri solely because you made a video about or else I would have never sought it out.
Bruh, if this kind of stuff surprises you, look up this four decades old thing called datalog, which was one of the oldest forms of the relational model. Modern implementations with different flavours include datomic and typedb. Vanilla SQL is just the implementation of the relational model that got big first, not the best one. You are just realizing that the actual relational model is really, really good, and that the main advantage of SQL compared to something better is being a universal standard. Having this kind of extended feature set where you can effectively do basic logic programming is usually called having a deductive database.
Since the creators of SurrealDB are actually on touch with you, can it be possible to launch a course about this one? A collab between you two, which could be actually useful for us as developers, gives them much more exposure for their product, and you actually get to earn some money monetizing the tutorials :))))))))
Chances are knowledge of this database has reached more people than knowledge of this guys RUclips channel. And there’s documentation so why would you need a course? That’s the give me the answers I don’t want to put in the effort type mindset
@@dabbopabblo nearly every piece of software out there that has a course for, is also documented. That point doesn't hold. It doesn't matter if other people know more than Jeff because some people really like the way that Jeff teaches and would rather have him doing the course, which is also a way to support a creator they like. Regarding your last sentence, why is that any of your business? It's not your time, nor your money. Why do you feel like this comment could be useful at all? Do you even know what are this person's intentions with knowing this tech? You know way too little to be placing judgment
@@YuriG03042 And what do you know about me to be judging my opinion? Its almost as if anyone can share any opinion they want and judging someone for being judgy is just hypocritical. Like your one to talk about not posting a comment because its "not useful" when you define criticizing an opinion as non useful, yet post a comment about your opinion on the matter. But my rebuttal to your counter argument in regards to my original point is; Sure general information tutorials and courses can be useful for understanding the fundamentals, however without digging into the documentation you will never know more about the subject than the person who made the course you took (assuming you only learn from that course and others) when the person who wrote or taught that had to learn from somewhere, and even if that was another course its not a chicken or the egg type question. The courses simply would not exist without the prior knowledge being available through documentation. Even in the case when none is provided because any project that lacks the industry standard of being properly documented will fail to be picked up by the mainstream and not be worthy of spending time making courses on. Point is if something has the potential to have a course made about it(disregarding general concepts of course), the information was already readily available and the people who actually take their work, hobby, or mere interest in programming seriously enough will already have the resources available to fulfill their desire to learn more about the product or library or framework in question with no course as crutches necessary. If a course is available as an appetizer into the content by all means necessary watching it can be refreshing before reading through potentially boring documentation. That's the only reason I watched this video as I haven't even touched this database yet and want a better idea of what its capable of before dissecting the documentation and refactoring my production code to use it. But in the end I truly believe requesting a course for something you have the means of learning yourself is a "I don’t want to put in the effort" type mindset.
@Batman Well if you ever find yourself using react maybe you should just reevaluate your choices and learn an actual programming language instead of syntactical abstractions on top of one, like all these kids who've been stumbling through RUclips having the audacity to refer to themselves as developers because they watched a few react tutorials have been doing
@Batman cause "react" developers are the kind of people on stack overflow admitting they avoided working with promises because they are "too" complicated
It's surreal 😍indeed! Your 100s series is good way to get introduced with techs in the industry. Whereas this format is great help for devs to decide to opt for a certain new tool, API or framework etc.
I used to write a roleplay website and struggled a lot with properly defining relationships between characters and ultimatively gave up on making it because of the complexity with parts of the implementation. This graph based querying solves so many problems its crazy. Love, love, love this!
Looks amazing, I just have to try it out now! I love this format of video even more than 100 seconds. Both are great though, 100seconds is just enough to learn what something does. Longer is better to get the basic understanding and some examples in. Great work!
'100 seconds' format is great, but so are the deep dives. Thanks to the superhuman information compression skills of the author I guess, I've seen deep dives that took an hour to say what you would say within 5 minutes. Easily. Thank you :) And Surreal seems like one of those cases where status quo of SQL/NoSQL has been around for so long that people got used to inconveniences, and only after you see what else is possible you go "wow, that's so much better, why no one thought of this before?".
Damn, surreal is awesome! One thing to note is, however, that it's not a full Backend as a Service like Firebase or Supabase. Meaning it does not have other features, like Authenti- oh no, wait, it has that. But it doesn't have Functi- oh wait, it has that as well. But real-time databases are still out of scop- Never mind, it has that too. You heard it folks, this is what BaaS should look like in my opinion. Proven concepts, expanded upon for more functionality, but not too much for it to be bloated.
I'm sure it's not going to perform as well as traditional SQL databases but for hobby projects the advantages of such a simple and versatile query syntax will far outweigh the performance cost. Really cool stuff!
..this beyond format is.. NUTS !! I love it. The perfect follow up to 100s.. period. Surreal looks awesome though. Awesome. Hope it will be FOSS anytime soon
Good mini-dive, like the additional video format. I kinda want to play around with SurrealDB myself but I need to think of some kind of project to use it in.
Please do a follow up video on the realtime aspect of SurrealDB. Maybe even go over how you'd approach realtime with a typical DB like Postgres or MySQL.
Holy shit, this is amazing! Time to start 100 more side-projects with this! JK, in reality, this fits a lot of my thought up use-cases so well. Hope they keep on their promises and wish the founders the very best in the years to come.
+1 on the use-cases. I have countless lists that I considered putting into a relational database. But always loathed the clunky setup for an SQL database. This has such an elegant syntax that I feel like finally realizing some of these lists that had been collecting dust.
@@dabbopabblo this may sound a bit vague, but many of them started from a creative framework where i tried mapping the existing categories to attributes from other areas. So I usually ended up with long tables that, when normalized, should be split up into a dozen separate tables.
@@dabbopabblo So in some regards, the lists are mostly static since they are mostly about conceptional groups than instances of user data or inventory. But things would definitely be less clunky if I were to create separate views with subsets of normalized data, rather than the big spreadsheet mess I'm working with now, lol.
Looks amazing. I like their model. They're doing it exactly right, for themselves and the community. Run it as a service for others, you pay for that. Otherwise, go for it. Perfect.I'm a little confused by the "identifier" syntax - coming from a background of integers and sequences and guids being used for identifiers - I love the "join" model (lack of need). I know I could just put a sequence or something there, but not seeing where they are headed with things like 'create table:row_id' - I think just getting back an identifier from the db like a typical db would do would make more sense. That said, I haven't dove into it yet.
The dot notation is amazing! I would be very interested to see the database performance for time series data, and the performance of the dot notation on tables with millions of records!
This is honestly kind of game-changing. Very rarely does something come along that offers something that is clearly a cut above the competition (looking at you JS frameworks). These mad lads took the best bits of all the major DB types and managed to mould them into a cohesive package. Would love to be able to use this in a real project someday, it's just a little too new to feel secure using on something that's going to be putting food on the table for not only your team but the clients. Hopefully, they can iterate on the performance side of things and get a solid foothold. I think good self-hosting will be key for this to be adopted, but I would totally be willing to pay a license fee for "advanced" features or get an enterprise support license, as long as there is a decent free version for hobby projects/testing/training. If this DB is thriving 5 years from now, I can 100% see this being the default choice when you start a new project.
The project seems really cool, but I find the Select thing with arrows not intuitive at all. Getting all the human who drove the car with this statement: " select
Feels like a middleground between SQL and Neo4j. But instead of having Cypher which is built for data with relations, you inherit syntax from SQL instead.
Moving on from C with zig, movin on from c++ with Rust, sure why not rid us of the join-hell that we've been putting ourselves through as well. Syntax-wise I'm all for this. Only thing that still needs to be proved is the stability, dependability and performance. But for personal hobby-projects I'll use this for - absolutely - sure, no matter the trade offs.
Been working with RDBMS for decades now and it’s all just so tiresome. Great balance of features and ease of use! This sounds amazing! Interested to see how it scales in a distributed environment and if it catches on in the market. Great vid, thanks!
I had already watched the 100 secs video and was looking forward to seeing this. Now my main problem is that I really really want to use it but have no project idea that could require a database. I am in videogames, not backend, after all.
The rust community is creating a lot of cool projects
this has been years in the making, and converted from a golang version into rust only very recently
not to deny your statement though lol
Interviewer: How are your programming skills?
Not Me: A little rusty...
Interviewer: Hired!
@@karpfenboy it only proves that rust is future and all other languages are merely the past
SurrealDB seems fairly easy to use, not to mention the joins. Keep coming with this type of videos. They are informative, long enough, and cool.
I've to say - videos in that format are very very useful to me. Enough information to form an opinion if a software is worth the time investing to learn more or even consider using it in a new tech stack. But the information is not too deep to get lost.Thank you, a very good video.
Seconded!
While the 100 seconds videos are always a joy to watch, I _sometimes_ get an impression that's quite different from the actual experience with the actual tool.
This is like the 101 that gives a look and feel of the thing in practice.
^This
Even if the performance gains don't arrive, it's still great. Not every project needs to handle millions of records. For projects that handle several thousand records the performance will be totally acceptable and the relationship syntax will greatly reduce development complexity. This seems like a must have in my toolbelt.
Definitely agree. Being able to write intuitive and easy-to-maintain code is also very important, especially if you only have a very limited number of concurrently active users.
Not sure I would agree here. It's a database - not scaling is not an option. I would store my data somewhere else if fast & synchronised access is not needed.
@@EdubSi again, it depends on the project,
True, I most likely don't need the best performing stuff, just need something to get it working , instead of spending precious time developing more complex and hard to maintain tech stacks.
I suppose if you're sure you'll never need to scale it, but most projects worth working on do have some scaling goals at some point. Not sure I'd want to invest in this technology then have to re-write it all in the future.
For anyone having problems following this quick tutorial (with Postman)....there are certain specific headers you should add (alongside basic auth)
In particular:
NS - {{namespace}}
DB - {{database}}
Accept - application/json
The Accept-app/json header is necessary with the beta-version...for some reason....but adding it will make it finally work
Man, I'm barely a developer and Surreal looks so goddamned awesome. I'm starting to think about a career change from architecture and if their promises hold up, I could definitely see myself getting deeper into this.
Same dude. Except I'm not an architect - I mainly sit around in my underpants watching RUclips videos, but this makes me consider giving all that up...
Think twice before jumping in development. Companies are laying off and the next decade might not be as bright as the last one. Cheers if you guys still make the switch :)
My man talks about changing his life around. I mean rn he sits in his underpants doin nothing and he thinks about developing, you tell him to think twice??? What did companies/colleague did to you man WTF?
@@ThunderboltPath This has always happened. Recruiting and laying off is just a cycle when it comes to IT and does not mean that things will get harder for future developers.
This is like clamining that "AI will steal our jobs". It just won't.
@JT wanna swap? Been a software engineer for 10+ years and writing code for 20. Wish I’d studied architecture and as building tangible things in the real world. Agree, SurrealDB looks Sick! 😍
The DB does look amazing, the queries seem so intuitive to write - I guess the only danger of it is that how easy it makes adding data that one might just no consider a proper structure before writing the code... But man does it look like magic...
You can also enforce schema too
that's like saying that spoons make soup so easy to eat that one might burn oneself, so better stick with forks.
It's also like saying that exposed wiring makes it so easy to do electrical repairs, but that one might accidentally shock themselves if they aren't careful.
Making it hard to foot-gun yourself isn't a bad thing in itself. It's at which cost we should argue about.
Some will prefer the freedom of schemaless. Others will prefer the structured nature of a schema.
I lean towards the latter.
@@pjarnfelt in your analogy choosing to use a fork instead of a spoon suggests it will take longer to program with a schema based db , however being burned represents your query’s being so in performant many fail and users are left waiting multiple seconds for anything to load on top of their own internets speed, and or crashes start happening from lack of ram to support the quantity of requests for how long each ones taking. And although as the developer you might not care about a little burn, programming is more like being a chef. Are you gonna serve your soup while it’s still boiling, or make sure it’s just hot? Cause you won’t have many users returning if their experience is being burned by a painstaking slow website or application because of non thought through database scheme or just lack of one
@@fluffio2976 If your application could kill someone (i.e. you're in critical systems) then you should be using an enforced schema and other safety checks.
99% of our applications are _not_ critical systems, yet our DB tools are designed as if they are. How many decades did it take for someone to design something intuitive like this?
My brother in law told me about SurrealDB last week. Having worked with ArangoDB before, its extremely similar in my opinion but with a future that looks very bright. Really liking what I'm seeing from it so far!
I love ArangoDB which is much further into development (mature) but this looks more intuitive! I'll definitely be following this project...it has that new, shiny vibe! LOL
You're by far my favorite youtuber so far, fun, quick and staying up to date what else a dev would want
It has basically everything you;d want which is amazing. I cannot believe how harder it will be to succeed though focusing on so many features. Will definitely use it in a future side project
Love the looks of this, and the beyond 100 seconds video format too. This is what modern DBs should be.
Modern DBs are what they are for a reason. Have fun diagnosing your performance issues on a production database after a dozen of developers have crafted the most obtuse and convoluted queries possible. The syntax might be simple on the surface but the generated queries under the hood are going to explode quickly
I’d be really curious about performance. If it can match a Postgres, I don’t see a reason to ever use traditional relational databases again. However I doubt it can if you really want to take advantage of those traversal methods.
It seems like it's built where you can pick the tradeoffs you are willing to live with. Some want "all teh speed" while others just want enough speed to be reasonable, but relatable code.
Here is a great reason for you: Traditional relational databases has ORMs, SurrealDB doesn't
Brilliant format. Short but also a quickest quickstart. Thank you!
Very much enjoy these beyond 100 seconds. Keep it up mate!
I'm currently learning back end development and this sounds amazing.
Would love to see some performance comparisons
Literally learning while taking a poop. Jeff always clutch with the Videos. 🔥
Very relatable
Yassss
Same
Same
@A english is sometimes weird… but then, so is the internet… 🥶
A video on EdgeDB would be nice
Perfect tutorial especially that it contains no music, thanks so much!
Love your videos. Especially long form. Thank you! The longer and more in depth the better.
Your videos are amazing. But the beyond format is getting another unreachable level. Just congrats!!!!!
Kudos to people who built this. I’m excited about it and would want to learn it’s internal workings
I’ve been working a lot with graphical db, and surrealDb seems just like that but more SQL like and with a such easy query langage, i love it !
I don't know why I want it so badly that this project to succeed!
Good luck guys! 💚
Man! A "deep dive" and it's only a little less than 8mins. You do amazing content and I know quality takes time but please moooore of this! I love this new channel idea btw!
You publish a video on this database I'm going to watch it and give it a thumbs up every time.
I love this format. You are keeping me relevant. Your videos are a great educational tool for continued learning. I'm building an app in tauri solely because you made a video about or else I would have never sought it out.
Who doesnt love going Beyond with Jeff! Keep this videos coming
Bruh, if this kind of stuff surprises you, look up this four decades old thing called datalog, which was one of the oldest forms of the relational model. Modern implementations with different flavours include datomic and typedb.
Vanilla SQL is just the implementation of the relational model that got big first, not the best one. You are just realizing that the actual relational model is really, really good, and that the main advantage of SQL compared to something better is being a universal standard. Having this kind of extended feature set where you can effectively do basic logic programming is usually called having a deductive database.
Since the creators of SurrealDB are actually on touch with you, can it be possible to launch a course about this one? A collab between you two, which could be actually useful for us as developers, gives them much more exposure for their product, and you actually get to earn some money monetizing the tutorials :))))))))
Chances are knowledge of this database has reached more people than knowledge of this guys RUclips channel. And there’s documentation so why would you need a course? That’s the give me the answers I don’t want to put in the effort type mindset
@@dabbopabblo nearly every piece of software out there that has a course for, is also documented. That point doesn't hold.
It doesn't matter if other people know more than Jeff because some people really like the way that Jeff teaches and would rather have him doing the course, which is also a way to support a creator they like.
Regarding your last sentence, why is that any of your business? It's not your time, nor your money. Why do you feel like this comment could be useful at all? Do you even know what are this person's intentions with knowing this tech? You know way too little to be placing judgment
@@YuriG03042 And what do you know about me to be judging my opinion? Its almost as if anyone can share any opinion they want and judging someone for being judgy is just hypocritical. Like your one to talk about not posting a comment because its "not useful" when you define criticizing an opinion as non useful, yet post a comment about your opinion on the matter.
But my rebuttal to your counter argument in regards to my original point is; Sure general information tutorials and courses can be useful for understanding the fundamentals, however without digging into the documentation you will never know more about the subject than the person who made the course you took (assuming you only learn from that course and others) when the person who wrote or taught that had to learn from somewhere, and even if that was another course its not a chicken or the egg type question. The courses simply would not exist without the prior knowledge being available through documentation. Even in the case when none is provided because any project that lacks the industry standard of being properly documented will fail to be picked up by the mainstream and not be worthy of spending time making courses on. Point is if something has the potential to have a course made about it(disregarding general concepts of course), the information was already readily available and the people who actually take their work, hobby, or mere interest in programming seriously enough will already have the resources available to fulfill their desire to learn more about the product or library or framework in question with no course as crutches necessary.
If a course is available as an appetizer into the content by all means necessary watching it can be refreshing before reading through potentially boring documentation. That's the only reason I watched this video as I haven't even touched this database yet and want a better idea of what its capable of before dissecting the documentation and refactoring my production code to use it. But in the end I truly believe requesting a course for something you have the means of learning yourself is a "I don’t want to put in the effort" type mindset.
@Batman Well if you ever find yourself using react maybe you should just reevaluate your choices and learn an actual programming language instead of syntactical abstractions on top of one, like all these kids who've been stumbling through RUclips having the audacity to refer to themselves as developers because they watched a few react tutorials have been doing
@Batman cause "react" developers are the kind of people on stack overflow admitting they avoided working with promises because they are "too" complicated
Finally! Some actual tutorials!
It's surreal 😍indeed! Your 100s series is good way to get introduced with techs in the industry. Whereas this format is great help for devs to decide to opt for a certain new tool, API or framework etc.
Man love the direct way you teach stuff.
Keep'n coming ^^
Love this style of videos for previews/overviews of new and upcoming things of interest
As a DBA this is awesome. I knew NewSQL/in-memory databases would be the future once I learned about it about 8 years ago.
I used to write a roleplay website and struggled a lot with properly defining relationships between characters and ultimatively gave up on making it because of the complexity with parts of the implementation. This graph based querying solves so many problems its crazy. Love, love, love this!
But it could be solved with a single SQL table. What was the problem?
so happy to have beyond 100s back
0:26, haha good primeagen reference there
I like this more than 100 second video format.
Man, SurrealDB sounds really great. I really hope that it rises in popularity.
Looks amazing, I just have to try it out now! I love this format of video even more than 100 seconds. Both are great though, 100seconds is just enough to learn what something does. Longer is better to get the basic understanding and some examples in. Great work!
This db is pure gold, and the beyond 100 seconds is great!
Halfway through this video I was like "shame you can't traverse back up the relationships though" and then by the end, I felt good!
Dude, you've become step #1 in chasing the techno dragon 🐉
'100 seconds' format is great, but so are the deep dives. Thanks to the superhuman information compression skills of the author I guess, I've seen deep dives that took an hour to say what you would say within 5 minutes. Easily. Thank you :)
And Surreal seems like one of those cases where status quo of SQL/NoSQL has been around for so long that people got used to inconveniences, and only after you see what else is possible you go "wow, that's so much better, why no one thought of this before?".
Because some pain in the ass must build up to make improvement worthwhile. 😃
I really like this more in depth content, keep it up.
This video and this video format is awesome. Please keep doing more of it
Damn, surreal is awesome!
One thing to note is, however, that it's not a full Backend as a Service like Firebase or Supabase.
Meaning it does not have other features, like
Authenti-
oh no, wait, it has that.
But it doesn't have Functi-
oh wait, it has that as well.
But real-time databases are still out of scop-
Never mind, it has that too.
You heard it folks, this is what BaaS should look like in my opinion. Proven concepts, expanded upon for more functionality, but not too much for it to be bloated.
Really enjoyed the beyond format, keep it up!
I like this beyond 100 second format
I'm sure it's not going to perform as well as traditional SQL databases but for hobby projects the advantages of such a simple and versatile query syntax will far outweigh the performance cost. Really cool stuff!
gogogo benchmark
What a time to be alive!
..this beyond format is.. NUTS !! I love it. The perfect follow up to 100s.. period. Surreal looks awesome though. Awesome. Hope it will be FOSS anytime soon
It's a great database, very intuitive and powerful.
Good mini-dive, like the additional video format. I kinda want to play around with SurrealDB myself but I need to think of some kind of project to use it in.
Please do a follow up video on the realtime aspect of SurrealDB. Maybe even go over how you'd approach realtime with a typical DB like Postgres or MySQL.
with the features it offers, the name is so fitting.
awesome job, it would be really interesting to know how this db works under the hood, multipurpose databases tend to not scale very well
Biggest takeaway for me was the ergonomics. Looks very developer-friendly.
I was not expecting Primeagen's blazingly fast mention :D
I have zero experience with SQL dbs, but this just feels so intuitive! Cool video, thanks!
Holy shit, this is amazing! Time to start 100 more side-projects with this! JK, in reality, this fits a lot of my thought up use-cases so well. Hope they keep on their promises and wish the founders the very best in the years to come.
+1 on the use-cases.
I have countless lists that I considered putting into a relational database. But always loathed the clunky setup for an SQL database.
This has such an elegant syntax that I feel like finally realizing some of these lists that had been collecting dust.
@@LinkEX what are you storing lists of thats static enough you weren’t using a database from the start?
@@LinkEX and if your lists are never changing why would you store it on a database instead of just a constant variable
@@dabbopabblo this may sound a bit vague, but many of them started from a creative framework where i tried mapping the existing categories to attributes from other areas. So I usually ended up with long tables that, when normalized, should be split up into a dozen separate tables.
@@dabbopabblo So in some regards, the lists are mostly static since they are mostly about conceptional groups than instances of user data or inventory.
But things would definitely be less clunky if I were to create separate views with subsets of normalized data, rather than the big spreadsheet mess I'm working with now, lol.
Please make video on your journey as a developer and the ups-downs you faced. btw love watching your video and it inspires a lot ^_^
Looks amazing. I like their model. They're doing it exactly right, for themselves and the community. Run it as a service for others, you pay for that. Otherwise, go for it. Perfect.I'm a little confused by the "identifier" syntax - coming from a background of integers and sequences and guids being used for identifiers - I love the "join" model (lack of need). I know I could just put a sequence or something there, but not seeing where they are headed with things like 'create table:row_id' - I think just getting back an identifier from the db like a typical db would do would make more sense. That said, I haven't dove into it yet.
Love the tutorial! I can see how easy it will be to build a content manager directly on top of this db
I like this beyond 100s format!
This is a good length to get a better understanding after your interest is peeked by the 100s vid
I'm super excited to try this out.
We like this format. Thank you and keep it up.
Thanks a lot for the tutorial! I'm thinking of trying it out in a small project I'm currently working on.
learned from this vid than all my 16 weeks of school lol
Great content presented with aplomb. More please. Thank you.
The dot notation is amazing! I would be very interested to see the database performance for time series data, and the performance of the dot notation on tables with millions of records!
This is honestly kind of game-changing. Very rarely does something come along that offers something that is clearly a cut above the competition (looking at you JS frameworks). These mad lads took the best bits of all the major DB types and managed to mould them into a cohesive package. Would love to be able to use this in a real project someday, it's just a little too new to feel secure using on something that's going to be putting food on the table for not only your team but the clients. Hopefully, they can iterate on the performance side of things and get a solid foothold. I think good self-hosting will be key for this to be adopted, but I would totally be willing to pay a license fee for "advanced" features or get an enterprise support license, as long as there is a decent free version for hobby projects/testing/training. If this DB is thriving 5 years from now, I can 100% see this being the default choice when you start a new project.
Super Video format and SurrealDB sounds great
Loved this format. Surreal is aptly named
I liked this "beyond 100 seconds" format, so I liked the video and I am letting you know in the comments.
SURREAL!
I found about about SurrealDB the other day, it looks incredibly interesting, can't wait until to use it for something
love the format!
I’d love to hear your views on Cloudfare’s R2 service !
Rhat DB is surreal, I am worrying about performance though... Concept and API is awesome!
I love this format.
I also really like the query syntax of arangoDB
Seems very interesting, curious to make some tests about this one!
Bro this seems like magic! 🤯🤯
Awesome! Thanks for another great video!
Wow, that's a lovely query language
Great stuff. Keep it up Jeff
Keep up the good work jeff ❤
The project seems really cool, but I find the Select thing with arrows not intuitive at all. Getting all the human who drove the car with this statement: " select
Yeah keep making these vids, I love them!
That looks fracking amazing
Feels like a middleground between SQL and Neo4j. But instead of having Cypher which is built for data with relations, you inherit syntax from SQL instead.
SurrealDB is to databases what Ruby on Rails was to web frameworks.
Moving on from C with zig, movin on from c++ with Rust, sure why not rid us of the join-hell that we've been putting ourselves through as well. Syntax-wise I'm all for this. Only thing that still needs to be proved is the stability, dependability and performance. But for personal hobby-projects I'll use this for - absolutely - sure, no matter the trade offs.
Been working with RDBMS for decades now and it’s all just so tiresome. Great balance of features and ease of use! This sounds amazing! Interested to see how it scales in a distributed environment and if it catches on in the market. Great vid, thanks!
Surreal seems like a fitting name by far
EdgeDB is beautiful
I had already watched the 100 secs video and was looking forward to seeing this. Now my main problem is that I really really want to use it but have no project idea that could require a database. I am in videogames, not backend, after all.
How about a shared progression system between games (like cod for example)
That's perfect though it could be a bit slower i think. Since this is a beyond channel, we can go in deeper slowly
Rust doesn't know the word slower