text stringlengths 0 2.35k |
|---|
[3959.24 --> 3961.30] you |
• Definition and purpose of immutable databases |
• Immutability in data and database context |
• Differences between keeping historical records and accessing current values |
• Use cases for immutable databases (e.g. tracking history, auditing) |
• Comparison with primary databases and traditional database use cases |
• Definition of immutability in databases, distinguishing between append-only data structures and verifiable databases |
• Explanation of how immutable databases use cryptography to ensure data integrity and detect tampering |
• Discussion of logical deletion vs. physical deletion in immutable databases |
• Importance of considering regulatory requirements such as GDPR when using immutable databases |
• Potential use cases for immutable databases, including financial transactions, health records, and sensitive information |
• The importance of immutability in data storage and verification |
• Benefits of using immutable databases, including simplicity and ease of use |
• History of immudb development, from cryptography and math theory to practical application |
• Dual modality for accessing data: key-value store and SQL database |
• Tampering detection and other types of verification |
• Possibility of querying data as it was in the past using temporary capabilities |
• Implementing SQL capabilities on top of a key-value database |
• The importance of immutability in package managers and databases |
• Challenges of adopting immutable systems due to human psychology (e.g. reluctance to admit mistakes) |
• Benefits of immutability, such as ensuring the integrity of transactions and providing a clear audit trail |
• Use cases for immudb, including software bill of materials and verifying the contents of software packages |
• Regulatory drivers for innovation in this space, including an executive order on cybersecurity |
• Discussion of SBOM (Software Bill of Materials) and its use case |
• Immutable ledger technology using immudb for storing and verifying software bill of materials |
• Protection against vulnerability exploitation through immutable storage and proof of authenticity |
• Comparison of running immudb to managing traditional RDBMS or key-value stores |
• Ease of deployment and operational procedures required for immudb |
• Choice of Go as the programming language for immudb development |
• Discussion of formatting styles in programming languages, specifically Golang and C++ |
• Unpopular opinions on working environments and productivity |
+ Jeronimo: Dislike for taking photos of working environments in scenic locations |
+ Jon Calhoun: Struggle to work in coffee shops due to neck strain |
+ Bartlomiej Święcki: Criticism of sedentary work habits and suggestion to incorporate physical activity into work, such as exercising while working on projects |
• Proposal for interactive and engaging ways to do work, including using standing desks and walking treadmills |
• Considering doing an episode on blockchain |
• Concerns about potential backlash or unpopularity of discussing blockchain |
• Polarizing nature of blockchain opinions and possible division among listeners |
• Goal to discuss blockchain objectively and raise merits if any exist |
• Fears of boycott or negative reaction from listeners |
**Johnny Boursiquot:** Welcome, everybody, to this episode of Go Time, where we get to talk about immutable databases. Now, I must admit, I'm gonna be a skeptic during this show, because I've been looking for use cases, and the project we're gonna be talking about does a very good job of articulating those things... Bu... |
Joining me today is my co-host, Jon Calhoun. Say hi, Jon. |
**Jon Calhoun:** Hey, Johnny. How are you? |
**Johnny Boursiquot:** I am good. Man, we haven't been on a podcast together for like a couple months at least. |
**Jon Calhoun:** It's been a little while. |
**Johnny Boursiquot:** It's been a minute, glad to have you with me here today. Also joining me are two -- I don't know if it's co-founders, or core contributors, or all of the above, but the two of you work on the Codenotary team. I did a quick google around and see that's a company that actually has a product that th... |
\[04:23\] Also joining Bart is Jeronimo Irazabal. Jeronimo also works at the Codenotary, on the team that works on immudb, and he's a software engineer, also passionate about cryptography and databases; I'm seeing a theme here... And also, he's been working on immudb actually a little bit longer, since the year before ... |
**Jeronimo Irazabal:** Hi, thanks for having me. |
**Bartlomiej Święcki:** Pleasure to be here. |
**Johnny Boursiquot:** Awesome, awesome. So first of all, I think our audience - not everybody is gonna be familiar with the concept; we all sort of share a common understanding for application developers, writing business applications and whatnot... We all have a pretty common understanding of your database. You write... |
**Bartlomiej Święcki:** Yeah, so when you have some information and you put them inside your database, usually we tend to think that this is some kind of temporal state, we can change this, alter this after some time... But what if actually there is some information that you don't want to change? That's where the immut... |
I remember when I was working on some standard databases, just common databases - there's this feeling when you delete too much records from the database, and suddenly you feel that "Uh-oh... How can I get out of this situation?" And immutability here helps a lot. It gives you this peace of mind. But there's also much ... |
**Johnny Boursiquot:** Right. Let me try to state that back to you, but based on the way I understand it. So when we talk about immutable data - let's just remove the database aspect of it for a second. When we talk about immutable data, we're talking about "What is the state of things? What is the reality of things ri... |
\[07:58\] So if currently it is 50 degrees Fahrenheit, at this hour, at this minute, and in another hour if the temperature rises about ten degrees, and now it's 60 degrees, you're not changing the past. You're not changing when it was 50. You're basically adding a new record, saying "Okay, another snapshot of this dat... |
So it's almost like you're dealing with sort of an append-only logging kind of situation, where at any given time you're able to go back in history to figure out what was the state of the world at this particular point in time. I can see why this creates some sort of a trail, a log, audibility, that kind of thing, and ... |
So it seems to me that immutable databases are about keeping history of things, not about being almost like your primary database. It's like, if I'm building a weather app, I may want to see what the historical value is, but if we change that a little bit and add, say, a financial services app or something - for a bank... |
**Bartlomiej Święcki:** In a sense, yes... Basically, this immutable database is like a watcher of a time. But it also contains the most recent state. If you want to check the balance, your current balance, it will still be inside this database. So there's still a use case as a primary source of information. But it's a... |
Let's have a use case where there is a banking application - I can simplify the use case - and there is a user. If you want to check your current balance, you open this application, check the balance. Then you do some purchases, and then you check the balance again. So you intuitively check if these things match. So if... |
And immutability also - and verifiability - can be used to actually make sure that not only the user can do this. You remember the old state, you know the current state, and you can somehow check if this is consistent. And immutability here, and especially in immudb, gives you cryptographical tools to make sure that ac... |
So if there's something crucial, like audit logs, which after some time you may want to do some investigation, what happened over time, this gives extra protection that you can rely on this information; because the database has proven that up until this point in time it is consistent with the whole history. |
**Johnny Boursiquot:** Okay. |
**Jeronimo Irazabal:** \[12:09\] I think you were covering a lot of problems that are addressed by immutable databases. First I would like to clarify, immutability is an overloaded term, because as, Johnny, you were mentioning, with immutability we usually refer to systems or data structures that are append-only. That ... |
So we are used to that for immutability. And actually, immudb relies on -- every component in immudb is an append-only data structure. Even the cryptographic data structure I would treat as append-only. But immutability in databases, or even in blockchain - we tend to refer to another thing; not just to append-only, bu... |
What we refer to this type of thing is verifiable. I prefer the term "verifiable database", rather than "immutable database." Because every system has integrity checks to check the consistency of a given record or of a given file, if it is consistent or not. But with tampering detection it's like, giving the possibilit... |
**Johnny Boursiquot:** Okay, let's pull on that thread a little bit... So we're not talking about the clients being or maintaining a copy of whatever data you might have at a central immutable database, or verifiable database. You're talking about some sort of a cryptographic verifiability of the data. |
**Jeronimo Irazabal:** So one of the particularities of an immutable database is that at every moment the complete state of the database is captured by your hash value. So that denotes not only the current state, but the complete history of changes up to that point. So the plan in immudb, for instance, or in other immu... |
**Jon Calhoun:** To make sure I understand this - that means that deleting records also isn't permissible? Is that true? |
**Jeronimo Irazabal:** \[16:15\] Deletion is actually -- we have two labels. We have logical deletion, or physical deletion. Logical deletion is something that can be handled by the application, or by the server. But the difference will be in terms of performance, because the filtering out of the information will be do... |
In immudb we currently have support for logical deletion in both manners - deleting a key, for instance, or by providing an expiration date. But this currently is just a logical deletion. This means that the data will be still there, it will be automatically filtered out, and the client won't receive it. But it's not y... |
**Jon Calhoun:** Yeah. And I'm assuming we're gonna wanna talk about good use cases for an immutable database... But I guess the first thing that comes to my mind is - I feel like you'd have to be careful as to what applications you use those for, because there are rules like GDPR where you have to be able to forget pe... |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.