I thought the German subs used more sophisticated encryption and were more disciplined in operational procedure such that the subs were largely not decrypted by the Allies.
I know my comment will sound too cynical or not well-received in a startup incubator forum. However, what I observed in common with the founders of the B2B startups I worked on was good networking.
The board members and initial contacts have much to do with the initial clients and the network effect it creates. It doesn’t matter if you have an excellent product. Without the initial boost of early adopters, you’ll struggle to survive. That’s why many founders go to SV or well-known startup incubators: the networking it creates is priceless. Of course, you’ll need to create a valuable product -and that’s the problem of many opportunistic startups leveraging current trends. Many startup pieces of advice I read seem to ignore the fact that you need money and contacts to have visibility in the right places (SV, elite university/circle).
While Paul Graham is a legend and I look forward to learning from his experiences, I believe that people in a position of privilege tend to overlook the advantages that come with their position.
For example, it is true that Steve Jobs started in a garage. However, both Apple founders were living in the epicenter of computing companies at the time. They both worked at HP and had contacts with people at Stanford. If you had built a personal computer in a garage in Latin America at that time, it would have been like a tree falling in the middle of nowhere.
Yes, you are right. The article discusses scaling. I got fixated on the concept of "founder mode" because the article invites a discussion about its meaning. My example involving Steve Jobs doesn’t apply to scaling, but without moving forward from the company's initial stages, there isn't anything to scale. I think initial networking also plays a vital role in decision-making during growth. For example, Steve Jobs' contacts with Xerox Parc greatly influenced the creation of Apple's Lisa, and Mac.
So, a book on "founder mode" could include multiple chapters on leveraging your network to scale your company.
This post brings back memories from when Windows dominated the market, and many online sites—e.g., Slashdot—talked about the "year of the Linux Desktop." (Sadly, in 2024, the closest thing to the year of the Linux desktop is Android.)
I tried multiple Linux DEs at the time. I also remember "Helix Gnome" (later renamed to Ximian).
I also remember that Helix Gnome used icons created by a person with the "Tigert" user alias. His icon designs were terrific and greatly influenced all the later Gnome icons.
The main Gnome problem at the time was stability: Nautilus and Helix Gnome generated core dumps constantly.
At that time, I switched to KDE for my daily work (I loved KDE's architectural consistency) and later to Xfce.
In retrospect, as a user of Linux DEs at that time—nowadays, I use macOS—the KDE vs. GNOME "fight" didn't help. KDE had a more stable code base but lacked a polished UX. GNOME had the investment of a few companies (like RedHat, Ximian, Sun, and later Canonical) that improved the visuals and UX over time, but its internals were a mess.
For years I'd go back between Windows and Linux, but sometime after 2018 or so I stopped even "going back" altogether. I've been able to daily Linux, my bar is high too. If it doesnt work OOTB I go back to Windows or to a distro that does. I'm not wasting any time fighting a broken OS.
I think the era of Linux on the Desktop is already here, just wish BestBuy would sell System76 towers, might push it further along.
Well, Linux as a desktop is quite popular amongst IT and developer types anyway. That's good enough fo rme.
In retrospect, I think I should be grateful that the Linux desktop never really went mainstream. It would have ended up being controlled by one (maybe two) companies and twisted into something that removed all of the benefit of it being open source to begin with.
I have basically had a history of using whichever desktop environment got in my way the least. At various times, that has been GNOME 2, MATE, XFCE, GNOME 3 (Ubuntu edition), GNOME 3 (PopOS edition), and probably a few that I'm forgetting.
Lately I have been running KDE on Debian and it is basically everything I always wanted Linux on the desktop to be for the last two decades.
It's wild how sane (and stable) linux desktop is today when you look back at some of the crazy experimental stuff that was going on 20 or so years ago.
In that thread, I mentioned the work on systems analysis and visualization done by the people behind Moose. (a framework for Pharo Smalltalk).
I didn’t follow that project for a while. It seems that their latest work is called Glamorous Toolkit, and the presentation is worth watching and relevant to this thread:
If you search for the keywords “program understanding” and “program visualization” in Google Scholar, you’ll see a lot of research on how to visualize and comprehend systems. Around 2007, one visualization that caught the internet’s attention was CodeCities by M. Lanza, and there is also Moose[1], which is based on Pharo and provides abstractions to query and visualize code bases.
But, for some reason, none of these tools catch the attention of the most popular IDEs. I hypothesize that there are major roadblocks to implementing this in a generic and useful way.
The first one is the variety of languages, frameworks, and build tools. For example, analyzing a TypeScript+React code base is not the same as analyzing a TypeScript+Vue code base, even when both use TypeScript and the TSC API is very easy to use.
The second roadblock is that useful visualizations also depend on the characteristics of your system, and creating them is not easy.
Maybe things will change with the addition of AI to analyze code bases, but so far, all the tools I’ve seen are either very niche or very short-lived.
It all depends on the requirements and quality attributes.
For example, a plain string template will not help with selective DOM updates; that’s why we end with a tree DSL like JSX.
For many different reasons, you’ll like to have lazy component rendering. That’s why using `Component(props)` is not the same as `factory(Component, props)`.
I’m not saying that the author is wrong or right. If you only want a server-side template or one-off template for a simple page, then using JS/TS string templates is better than relying on a DSL like Handlebars or Pug. The client-side story is a bit different, and that’s why there are so many frameworks for it.
> Your first leadership principle is customer obsession: “Leaders start with the customer and work backwards”.
> I’m your customer, and I’m begging you: please let me be a cloud engineer again.
However, as with many enterprise products, the author is not the customer; it’s the user.
The customers are the companies that buy AWS because it’s an essential technology for their strategy. When the whole tech world is talking about generative AI, they want to be there, and Azure seems to be ahead because of the MS deal with OpenAI. (even if they are not ahead, customers' perception matters most).
So basically, what Amazon is trying to do by making all of these conferences and announcements about GenAI is to send a message to their customers: we are ahead on the wave of GenAI and you can still trust that our products are going to help you be on the hype.
The goal is to use the best tool to solve a problem. But it is tricky when you don’t know better and everyone around you follows the herd for different reasons.
Most presentation tools are poorly designed.
For example, many cognitive studies support the idea that the same brain channel handles text and speech. If you try to read and listen to different things, you can’t simultaneously pay full attention to both. Yet most presentation software encourages using bullet points, which are used extensively in business presentations.
Superfluous animation is distracting, yet presentation software animation capabilities are redundant and don’t have basic things like having a timeline with keyframes.
Outlining and notes are handy during the ideation phase of a presentation, yet outlining features are terrible.
Google Slides is extremely painful to work with. Taking a template and filling in the blanks with bullet points is okay, but even formatting a text box right is painful. PPT is better, but it has the typical MS problem: feature bloat, where the most valuable things are half-baked.
Keynote is one of the best. However, its sharing editing capabilities are between buggy and nonexistent.
By contrast, Figma's visual editing capabilities are orders of magnitude better: components and auto-layout help you work faster with a tool abstraction that is much better than the crappy master slides.
So, if you work the whole day with Figma and are productive with it… why inflect yourself the pain of using PPT/Google Slides for your presentation? You will be the one presenting. (Note: many times, I used Google Slides for those shared presentations, which I suffer… but if its my presentation I prefer Keynote or Figma depending on the case)
Slides are consumed in a variety of different ways.
Slides can be something to keep the eyes busy while someone is talking IRL. Alternatively, they can be a way to exchange information outside of a presentation, reading them like a book. In the latter case, bullet points are quite useful.
A lot of features of presentation software are there solely to stop people from going to sleep.
The problem with XML is that most tools around 2000-2010 used it for everything, including simple configuration files and even DSLs (the Maven v1 XML was a nightmare to maintain).
However, there are features of XML that are very useful for a document format like SVG. One of them is namespaces. For example, you can create an SVG and add custom metadata in another namespace, and then any viewer who uses a decent XML parser will show it without issues. That’s not possible with JSON, as there is no guarantee that an unknown key will pass. (although I prefer a simple format like JSON or TOML for config files)
https://en.m.wikipedia.org/wiki/Colossus_computer
reply