hackmas2026 - Basics of Personal Information Management: Finding the best tool(s)

So this is a a little bit of a a mix. It's not that um like like my usual talks with one single topic. So it's more of a broad uh overview of a certain aspects of personal information management. Okay. So let's start um first of all which what what is the scope of this talk? So as I already tried to explain uh it's personal information management. It covers a few basics of this topic and the goal is to enabling you all uh to make better uh choices for your particular situation. Yeah. And that's me uh here down in the corner. Uh I enjoy very much looking other people over the shoulder when they're working with computers. Uh and uh if they want I sometimes may have good advice for them in order to get better to some degree. Not all people wants that though. So I had to learn the hard way that there are some people that are actually quite angry if you want to provide them help with their work and their workflows which is a little bit a little bit confusing for myself. Um the next thing we want we need to talk about is what's the definition of um personal information management when you hear it from me. Uh there are usually two different definitions of personal information management. The one the narrow one is uh calendar to-do contact on uh management uh all the way like like outlook or Thunderbird to some degree and so forth. And the other definition is the definition I came up with is um how you organize yourself and your data. So it's a very very broad spectrum which I think is personal information management when I do it. uh and of course uh there is also uh the non-personal part which is very similar. Uh so we have uh requirements and methods on both sides and um the group tools and the group workflow implementations of course are a little bit of a different turf than the things I usually talk about but we have certain exchange of ideas and tools. So it overlaps a little bit but most of the time I'm concentrating on the left hand side personal information management because one brain is complex enough and this is uh supported by the terms of different mental or cognitive models. Some say uh some prefer the term mental some prefer the term cognitive models. So which is just a fancy name for uh you have a brain other people have brains and they those brains differ from various for various reasons of course and another thing which is very important here is that your own brain changes over time. So the situation you have been 10 years ago for example most probably was a different one different age different job working area whatever. Uh and so if you are working with information storage and retrieval you need to take into account that the retrieval process will be with another version of your brain in the future which is tricky. Okay. because you have to somehow see what's going on in the future which is mostly difficult task to do. Yeah. Another thing I wanted to u talk about briefly briefly is uh the desktop metaphor. The desktop metaphor is uh a very very cool tool. It was developed in the 1970s um by some research uh re researchers in uh at the Xerox Park Institute for example. Um and this is one of the earliest screenshot of a graphical user interface uh with the desktop Metafera as we know it. You see there are icons, there are printers, um uh there are documents with icons and if you would have moved such a document over to this printer for example, you could have um you could print this document um at the network connected printer which was extremely sophisticated for the time it was developed because the alternative would have been text consoles. Uh and with the desktop metaphor, it was the first time that you could sit any person in front of a computer with no technological background at all and assume that this person is able to work with the computer. And this was very new back then. And of course the trick that was used for the desktop metad was that uh people could transfer the world knowledge to the virtual world and also inherit. Yeah. And the bad thing is that it also inherits restrictions from the phys physical world which is a bad thing because in the physical world for example each and every item has exactly one single physical position. In the virtual world this is a restriction which is actually not necessary at all but we still chose to keep it that way. Okay. So therefore I warn people for uh using too much of these old tricks like the desktop metaphor uh because it's some kind of limitation for our modern world as well and those limitations hurt more than the uh trick the positive aspect of the desktop metadata actually helped us back then. Okay. I always uh use the the the the um the phrase it's like uh running a marathon with uh uh chained hands for example. Yeah, it's possible but it's no fun at all. Okay. And we are running uh with uh jane hands all day long and everybody seems to be happy with that situation. And sometimes I think that I see that there is something you might have uh potential to improve and people do not want that or do not see the problem at hand. Another aspect you everybody of us including me perfectly knows well is naming things. Naming things is hard. Uh and we we do we do name things all the time. We do name things when we define document titles, headings within a document or folder names. This is a very uh interesting example because this is new folder one two three is probably the most frequently used folder names ever. Uh file names uh button labels. So this is uh more or less the domain of programmers. Uh menu entries you see the the p the pattern. And why is this so difficult? And there is a very interesting research work um for example 1987 from Mr. Furnus at all uh which was called the vocabulary problem. Who knows the the this research uh or the term vocabulary problem? Nobody. Okay. No problem. Quickly explained. Uh there was an experiment. Uh imagine you're entering as a test person. You're entering a room and there is a big table and the test persons sit around the big table and uh the researchers put a thing in the middle of the table. It could be as simple as this one. Okay, this one just placed on top of the table and then the participants are asked to write silently write down the first word or the first term which came to their mind what this thing actually is. Then they write write the term down. They write the the word down and then the researcher said okay now let's forget your first word. for some reason this is not valid anymore or not interesting or not matching whatever. Write down the second word or term which comes to your mind what this thing actually describes or is. And this goes on until I think five uh five levels of of terms and words. And of course the first one is the most obvious for yourself. The second one is the second most obvious for yourself and so forth. And the interesting part is that there is almost no match among the participants in the first round. So there is no agreement on such a simple thing like a physical thing. And it even gets harder when there is no physical thing but some abstract thing. Yeah. like a program function. Usually this is very abstract. Okay. And you if you need to come up with a term for a a feature, a virtual feature in a software, it's even more it's even harder like that. So what the researchers did is they looked how many levels of words they need to go down in order to find one word that was mentioned by everybody or the majority. And I I can't remember the details but the basic thing is that you have to go down uh in that hierarchy very deep in order to find one single word we all would agree with. Okay, probably it's bottle or water or I don't know beverage whatever. Okay, in this case it's a very simple example, but you can imagine that if the the thing I put on the table is not as simple as this bottle, this gets trickier and trickier. So the basic idea or the basic outcome of this vocabulary problem research is that different people relate different words uh to the same concept because they have uh different concepts in their brain. They have different experience. they they see it from a different angle maybe and therefore it's very difficult to or it's it's it's it's not good to assume that everybody thinks the same way I do or that everybody thinks that this thing has the same name that I would use to describe it. And when we go back to the naming things slide, you see that document titles, well, that's not that hard because document titles usually usually are very descriptive and longer. It gets harder for example with folder names and file names because people usually tend to keep it short and when you keep it short you have to probably uh select only one single word and your knowledge of that one single word is a different kind of cognitive model than other people's have for the same folder and the same content. You get the idea. So therefore the vocabulary problem is one of the basic uh informations that we got from PIM research which tells us that naming things is a very hard thing to do actually and which leads us to the next big chapter for information storage and retrieval and if we talk about storage and retrieval it's very important to emphasize right from the start that you don't need to concentrate on storage when there is no retrieval at all. People tend to concentrate on the storage process although they should think of the retrieval process in future. So you should put your mind in the state where you assume what will I be thinking of when I want to access this information in future again it's even harder when there are multiple persons involved of course. So we only store information because we want to maximal uh maximize retrieval success when we need that information in future and when there is no need for retrieval there is no need for storage. The hard thing is that we all want to store things and we do not know if and when we want to retrieve that information again. That's a different problem. And when it comes to storing information, there are a couple of uh things that modern PIM research has come up with. Uh the classical categories of uh people filing stuff is filer, piler and string cleaner. So filer are people that um store away documents neatly in their uh in their folders and so forth. Pilers are people that have huge stacks of paper on the desktops or on the virtual desktops on their computer. uh or few folders and a little bit more cha chaotic some would say and spring cleaners are the classical mixture. So usually the people that are piler uh become a filer once a year or so when they move away the things and delete the other things and so forth. This is this mixture is called spring cleaner. But reality is more complicated than that. While this is basically true, there are all kinds of different mixtures. Even in different situations, people tend to behave differently. So for example, with paper, people tend to be for example a piler on the computer. They want to have a neat desktop and so forth. They they tend to uh behave like a filer, for example. Everybody's different. And of course then there are decisions like keep everything versus delete as much as possible. So there are people that want to get rid of everything right up from the phone. Now I don't want to store that. I if I read it again I most probably will will refind it on the internet and so forth. Okay, that that's one uh possibility. Yeah, it's not worse or more correct or whatever compared to the other. It's just uh and different approach. And of course there is this order versus chaos. Some people are quite happy with something other people would say is a chaotic situations. Usually people maintaining such chaotic situations do not think of themselves to have a chaotic situation because they somehow come around with their workflows and processes. They they know how to refine their documents because somehow they can manage this chaos. Other people need more guidance. Usually these people tend to be um uh filer and not piler and so forth. Uh when it comes to uh working with computers, we have uh the need for storing um computer files locally. And uh just for the sake of this talk, I want to make sure that directory and folder is the same term here. I I use them uh in the same way and the same holds true for file and document. Of course in details there are differences but let's put that aside. I already mentioned the desktop metaphor and that we inherited downsides from reality in the in the virtual world and the same holds true for so-called modern file systems because even in our modern file systems it's the case that one file is stored at exactly one single location in the file system although the computer could do it differently for example like in a database system and so forth But our modern file systems still do not have the same features as a database. So if you want to have a file on multiple different locations in your file uh system hierarchy, you need to use workarounds and all come with different disadvantages and most people don't even know about those workarounds other than copying files and so forth. Uh again I think that's a needless limitation. We took from the real world into the virtual world and we shouldn't do that but it's very hard to change. So uh one thing that relates to the previous one is that uh I think that there's a issue if you want to transfer something from the real world to the virtual world. So uh you can't map reality to a strict hierarchy for example. People try to do it all the time. People try to come up with the perfect folder hierarchy because of that reason or the perfect mail folder hierarchy and so forth. Okay, that's not possible. That's only possible to a certain degree. If you can deal with the negative aspects within your brain, within your cognitive model, then it's it's perfectly fine. For example, I don't see an issue with that. But or but my brain is probably different from the average brain in that respect because I very rarely has an issue refinding uh an information I stored somewhere. But that's not true for the general public. There have been uh several approaches to uh come up with some standard way of using a hierarchy. For example, the due decimal classification is very uh famous. It was I think the origins are from uh librarian uh and so forth. Uh but all of them are horrible to use in my opinion and not a very good idea especially if you want to make use of the advantages of a computer. Um, yeah, and all of those are rooted in the need to order physical items, which we should not do in a in a virtual environment, but we do it anyway because we learned it that way. And nobody seems to care that actually we could have better opportunities, better possibilities, better workflows than that. I'm using an allegory here uh which um compares buying a digital camera from a physical store, a brick and mortar store uh and uh buying a digital camera in an arbitrary online store. When you go to a physical store, you um depend on finding the area within the store where digital cameras are exposed. You can choose one of them. you can take it with you, go to the um point of sale and take it with you. And that's the way you buy digital cameras in the physical store. Everybody knows that in an online store, it's a completely different process because when you visit an online store, you may find the very same thing, the very same digital camera on 10 different places. It might be on the front page. It might be in the office supply area. It might be in the uh outdoor equipment. It might be at the children toys area. It might be catch of the day. It might be, of course, in the digital camera section. and nobody gets offended that the same camera is linked everywhere which it might be okay to to look for that camera. So this is this is the good case. Okay. And for some reason uh with our local computers we are find that we just want to have the same physical store methods we using with our so-called modern operating systems which are actually not so modern if you think of it. So if you somehow need to do file management with hierarchies, I do have some recommendations, some personal recommendations. Yeah. Uh my talk is half uh based on research and half based on my personal experience. Uh I did a PhD with file management um 15 years ago. So I hope I know what I'm talking about. If you need to maintain file hierarchies, I would recommend to keep it as minimals as possible. I once was world champion in coming up with very complex file system hierarchies and they all failed within a couple of years. Um, keep folder names as generic as possible except the last level. Okay. So maintain some or stick to general concepts until you come to the second or third level in your file system hierarchy and then go to the specific uh specifics for example on file system uh or on file level and not on folder level. Ignore. Please do ignore those stupid default browsers. Uh modern operating systems like Windows and Mac OS come with for example my document, my pictures, my videos, they do not make any sense at all. Yet there are standard site since decades. Okay? Because if if I'm joining for example a party, then the invitation would be a PDF that goes to my documents. The photos I take would be my pictures for the same for the same event. And if I happen to make a short video, then I uh would uh be um I would be um guided to to use that folder for the video. That doesn't make any sense uh sense at all in my opinion. And the bad thing is that when people do not came to my talks or my lecture or read my books or my blog or whatever, they think that this is actually a good thing because Microsoft is telling me or Apple is telling me, uh, yeah, these are the default folders. Please use it. For some reason, they come with neat icons and so forth. So people somehow start a horrible organization because of those stupid defaults. So I I would say avoid numbers in file names. Um like 00 important stuff and so forth. Uh it's it's it would be be a talk in itself on the downsides on the subtle or not so subtle downsides of adding uh numbers in file names and folder names. So uh this is my general recommendation and if you are using multiple hierarchies in parallel for example for your local files for your emails for your bookmarks whatever then I would urge you to stick to the same principles or the same hierarchy at all. Yeah. uh if you are looking for different options than hierarchies um I there is probably one thing I want to tell you here it's from a book called miscellane everything is miscellaneous from David Weinberger highly recommended for everybody it's not a research book it's very easy to read and it covers all types of uh cool background information from the past uh how things developed as they are and uh he's talking about three orders of orders. And the first order of order is ordering the items themselves. So the books in your shelf, you can order them by color, by author, by title, whatever, by age, I don't know. The second order of order is some index card systems like in the libraries where you have let's say three different cards for each item and you can order the cards any way you like. So you have in parallel the cards in one in one uh cabinet the cards are ordered like uh from the author names the other ones uses the title and the third one uses the uh date or something like that. Okay, fun fact. Most librarians would say that color would be a very clever thing to order books because many people come to the bookstore and say, "I want to buy that book. I can't remember the title or the author, but it was green." And it's that's a very uh very common issue they they have to face. And the new thing here is the third order of order. So, ignore the other orders. use the content, split it up in digital chunks and use a method that enables you to use that chunks to refind the information on demand. So you decouple the situation from storing to the situation from refinding or retrieval uh in that way that for example you're using search or in another situation you want to su you want to look for the author and so forth. So it doesn't matter which situation you're in when retrieval is is your thing. You can come up come up with different workflows and processes and tools in order to refind the information any way you like. And of course for that you need a computer. That's not possible in the physical world anymore. Yeah. Which leads us to the next uh topic I want to talk about briefly. Uh multiclassification. Uh multiclass classification is just a term uh which you might know with other terms like tagging or labeling and so forth. So you assign keywords or keyphrases to information as a meta information on information. Uh so one thing can have one to many different associations labels tags uh to different concepts. So those labels and tags refer to some kind of concept you have in your mind. Uh, and although most people think that this is a very straightforward task to do, I can tell you that it's not because if you do not keep a certain method or workflow, those labels won't help you very much in refinding the information. uh from my personal experience as well as from the research I've read and as well from my PhD I wrote about this topic, I came up with uh 10 rules you might want to keep in mind if you are following labeling or tagging or whatever you call it. So I would suggest to use as few tags as possible. So if one person is involved, social tagging is a different turf. Okay, I'm talking about one person tagging his or her own files, for example, or web pages or bookmarks or whatever. So, use as few tags as possible. How many depends on your situation. Uh, limit yourself to a self-defined set of tags. That's of course in combination with the first one. So, you should refrain from coming up with new tags with every storage process. So if you have used for example education uh for labeling some education stuff, you should not use school or university at the next storage process uh because then you have two different and very similar concepts for the same type of information. This is often referred to as a uh controlled vocabulary. I've developed some file management tools which are providing uh features for controlled vocabularies. Of course, um tags within your tax set must not overlap. So you should not for example use sports and swimming because swimming is a subconcept of sports with some exceptions. Yeah. If you like sports in general and you are very much particular interested in swimming because you're there uh for your job or you're in a in a far was um in some swimming association whatever then you might use both tags. Okay. But in the general case you should refrain from using tags that overlap whose concepts overlap. By convention, it doesn't matter if it's plural or singular, but you have to stick to either one. Yeah, in my case, I use plural. You can you can choose different. Uh, and for the same reason, lower case, for the same reason, single words because multiple tools have different approaches when it comes to um differenting or finding out the end of tags and so forth. So, I would stick to single words. Um, keep tags on a general level. omit tags that are very obvious. Yeah. So, you probably won't or should not use the label picture because you might get that information from the context, the file name or the file extension. Uh, use one tag language so you do not have to remember was it school or schul in German. Uh, and explain your tags. keep a text file where when you add a new tag to your controlled vocabulary, you explain it to yourself because I found out that I introduced tax in a couple of years uh after I was not sure how I meant the tag in which context what is the specific context and so forth. So explain your text to yourself or your future self and uh yeah here I is a brief example of useful text you might get the uh ideas from for example selection is something I'm using for my photographs when I have let's say from that uh Hawkmas event uh I return with 200 photographs and I don't want to show my wife 200 photographs then I might have selected 17 of them with the tag selection and then I have a shortcut which only shows the uh files that contain that tag and this is very easily presented then um taxes if you do taxes like me once a year uh over the year you collect bills and some of them some of them are uh um some of them are related to uh taxes then you might uh tag your uh files that relates to that over the year with the tag taxes and then it's very easy to relo to refind them. Bill oops bills for the same reason. Manuals uh I I love the tag because manuals come in as uh word files as PDF files as screenshots and so forth. Yeah, you get the idea. screenshots. Uh then you can use text for example for uh driving workflows uh like on the uh bottom here draft submitted and published or public confidential secret if you want to assign confidential information uh the the tag of confidential information to a file and so forth. When doing general file management, I would also um u give you the advice to keep a to delete folder somehow which is your default download folder and just because of that name to delete you know that you can remove all content anytime. So either you file away the content if it's important to you or it's something that you can throw away any time. Yeah, dedicated inboxes. Uh yeah, here are some examples for folder names. I I do not go into the detail because we have an time issue I think. Uh but you can get the information afterward uh on the slides and I have a blog article with more details on that. Uh the basic idea is that uh there are some bad examples of of file names uh and there are some more helpful file names and the the bottom one for example is using the draft tag uh instead of version two because those version two information for example doesn't come with any information at all because yeah version two of how many I don't know was it a draft is it the final version but if you make it explicit like for example example with a draft tag you [snorts] know okay this file is in the status draft and not finalized or submitted or whatever. [snorts] Um here is an example of my personal file management method which consists of a file name convention. You see, I usually start files with a date or time stamp, then with um an an a normal file name, you would say, and then optionally zero, one or more tags, which are called file tags. And there is this space minus minus space uh divider between the normal file name and the text. And I developed a set of tools which are able to deal with them. So for example, moving away files in their u in their folder I do most of the time with my tool called move to archive. Uh adding date and timestamps I do not do it by hand. I use date to name. Um for that management I use the tool append file name. And managing many many different u features for those file ts I'm using my tool called file tags. uh you may want to check out the information on the last slide then where there are links to um further articles on my file management method. So now that we have covered some topics on the file storage uh area I want to briefly mention some things uh related to information retrieval because basically there are two different uh methods to retrieve information. one is uh search and the other is navigation. We all know this research shows that search is the dominant uh um retrieval process in the internet on the internet in the file in the in the in the uh web browser and navigation is preferred by users for local file retrieval somehow. Uh I would urge you to use both on any situation. So uh when it comes to uh local file retrieval, you should also be able to use local file search because sometimes it's hard to refind the folder location of a specific information and it's easier to just to use local file search. And there are other mixtures and other retrieval methods for example like faceted search or tags which is my uh my domain or teleporting which helps refinding information on your computer. My recommendation not focusing on one. You should have a basic idea of all of them and you should come up with a feeling which retrieval method is the best in my current situation and this is a different one today than for example in two weeks from now. The next thing I want to emphasize is that um uh modern computers do have way more possibilities and and file retrieval u uh methods than we uh used with the default file browsers that come with our operating systems. Uh so don't forget file search. I already briefly mentioned create books marks for frequently visited folders or use a freency. And this is not a typo. Frey is a mix of recency and frequent frequent yeah frequent. Uh so uh for example when I want to visit a folder I use a tool called set like the letter set and then a search term and then I then it jumps me to the right next um suitable uh folder name which refers to that search phrase and this is very handy. So I rarely use CD blah blah blah in my case although I'm working in the in the Linux uh terminal all the time. Uh Windows file explorer Mac OS finder are really limited tools. I would not suggest to use them at all. Maybe play around with alternatives. Uh for example free commander for Windows or this is a very interesting one Fman uh which is from a guy in Vienna. uh just even if you do not want to use it, take a look at it because it comes with certain um uh features that are very handy. for example, that frequency uh method uh I found the first time here in in FM. And if you want to for example uh look for uh large files that um make uh that that um you want to get rid of because uh your free space uh is is is uh very limited then I would say you should look for a tree map visualizations tool uh and this looks like that for example. So the the larger the area here the larger the content of the folder or the file whatever other topics of course which I cannot cover here today millions of other topics but you should um use proper backup uh you may be interested in file synchronization that is not using the public cloud uh but your own computers for example like syncing.net net or nextcloud and so forth. Um, and don't be um intimidated by me. Whatever floats your boat is okay with you. Okay? So, you must not change anything at all if your concepts are working. However, many people's are lying to themselves when it comes to is it really working? So, don't lie to yourself. Reflect your situation from time to time. things change and when things change your process might need a change. And uh with the last uh couple of minutes I want to go briefly to my um proposed workflow flow of choosing the right tool for the job. So with the basic information from some parts of the research of uh personal information management, you should now have um much better ability to choose tools that suits your personal situation. And this is different from every one of you. So whatever you you're choosing might not be the a good idea whatever you tools you are choosing because your situations differ. And this is my proposed workflow. Requirement analysis, choosing a method, analyze the tools, choose a tool, and then optimization. [cough] So what [clears throat] do I mean with requirement analysis? So in the in the simplest form, you do a short brainstorming. You sit down with a white sheet of paper and write down whatever comes to your mind. What do you actually need? not what do you want, what do you need, which is different sometimes from what I want. Uh create a list afterwards after the brainstorming. What of those things you collected in the brainstorming has priority to you? What is a a must-have? What is a nice to have? And what are explicit non-goals? Very important non- goals. Uh and you should not uh forget to include topics like flexibility to adapt to changing environments because the more flexible your tools are the more likely is that changing situations do not force you to change the tool. Uh you should define your lockin appetite. If you need a tool for certain weeks or months then you probably do not need that. But if you want to keep that tool for years or decades, the lockin situation here becomes more and more important for your future self. Uh and lock in means that uh uh I there are some migration cost associated after end of life of your solution which I'm going to use now. Data protection, security, privacy, dependencies and so forth. You know the drill. So then you should choose a tool uh you should choose a method and uh methods are not always there for certain aspects. For some aspects it's very easy. For example, when it comes to mail email management, you might stick with a method like zero inbox or chaotic inbox or whatever. For file manage management, you could stick to one folder with everything and I want to have nice search or you would you want a tool which supports a deeply nested hierarchy of folders and so forth. So whenever it comes to um method versus tools on the left hand side there is an example of some um general method and on the right hand side you see an example for add tool. And here is where most people fail because they start with a tool and then look for methods or they ignore methods at all. So identifying methods it's not that easy sometimes. So therefore I would urge you to search the web, ask peers, uh read books. Ask peers is a quite cool example here on this hakmos. uh you might uh meet many people here with similar requirements and you can exchange ideas. Read books. Uh match the known methods to your prioritized requirements. And this is very important that you have a prior prioritized list of requirements from the first step. You remember by doing so you learn from the pros. You do not have to make any every mistake by yourself. Please spare yourself some lifetime and profit from the experience from others. And notice that still at that at that stage we are still independent of any specific technological solution or tool. Okay. Still no tool up till now. Now analyze the tools on the market. So again, search the web, ask peers, read books, and don't forget to include small, new, and unusual solutions. So look around and maybe there is a newcomer on the market which might be a very cool uh solution for your ideas uh but it's not that popular and you would overlook it if you don't look specifically for small or new ideas. That would be a pity. Don't focus on tools that are already known to you only. Uh never underestimated migration cost locking effect. Migration C that results from overlooked lock in effects. This is very important because most of the time people totally ignore lock in effects and then they're using tools where they can't export their family pictures anymore like Apple did many years or they're using a personal information management like one note where there's almost no way of exporting your data or Obsidian and so forth. It's very hard for people um it seems because many people uh underestimate that negative aspects and you need to face it when you move away to the next tool in future and then choose a tool. Okay, with all the work we did so far now we are in the position of choosing a tool. So match the available tools to your prioritized requirements. You may assign numbers to come up with some sum and indexes and so forth. Um, now is the time to profit from your preparations in this process so far. Make sure that all musthaves are fulfilled. Don't overlook what you asked for in the first step. Don't be intimidated by the new shiny tool and forget all your requirements. You need to match your requirement because that is the list of things you want to accomplish, not to use a shiny new tool which might not be so cool anymore in a couple of months. For example, and retrospectively modify your requirements only with very good reasons. So for example, if you're looking around on the tool market, you might come to more stumble on more ideas that may be a good fit for your processes, but you have not thought about that in the first brainstorming session. Then it might be okay to add requirements retrospectively. But do not do it for no reason. Okay? It has to be very very good. um thought of yeah and then the last point is optimization. So uh periodically improvements and re-evaluation of your requirements and choices as I already mentioned your world changes, your situation changes, your business changes, um your hobbies change, maybe you found a family or whatever. Yeah, your situation is a different one in a couple of years than now. And therefore, re-evaluating your past choices from time to time is a good thing instead of sticking to a tool which adds more and more burden to your daily business. Yeah. And this is the basic how to choose a tool process I would propose to you. You see that's not rocket science. Nothing from that is based on my own experience only. It's a no-brainer. Basically, the important thing is not to forget anything is in this process. It's like a checklist and you should tick all the boxes before you make the decision because shiny new tools tend to attract people and tend to let them forget their own requirements or overlook the lockin effect. the tools come with. Okay, so I hope this was not too fast, too much or um did not meet your expectations. If you want to find out more background information on the topics, you might scan this QR code or visit the URL of the QR code. Uh it's a blog article on my web page with couple of links to other resources. Uh and uh with that I would ask for some questions uh if there are we have a couple of minutes left not much but I'm staying here of course and you can ask me anything you like all day long I'm here until Sunday. [laughter] So any quick questions now? Okay then we'll do it in the break afterwards. Uh thank you for uh coming here. Thank you for your interest in my work and you find much more content on my blog and uh I have tons of videos online from other talks. I'm I'm sure you can uh find yourself uh additional information to certain aspects of that and if you somehow related to university I do have a lecture which I held at uh Fasank Burton and Gats University of Technology and I would be happy to continue that at another institution as well. Thank you. [applause]