Hacker Newsnew | past | comments | ask | show | jobs | submit | nl's commentslogin

Fraktur actually does use a partially different alphabet. For example it uses the Long s: https://en.wikipedia.org/wiki/Long_s and Half-r: https://en.wikipedia.org/wiki/R_rotunda

I rather assumed so as well, but a big of digging turns up a whole history: https://en.wikipedia.org/wiki/Antiqua%E2%80%93Fraktur_disput...

Surprisingly to me the Fraktur typeface was the traditional "German" typeface but was disliked by Hitler.


Does anyone remember the Yahoo design patterns library? It was mostly for UX pattern (eg: ways to "Rate an object") and it was really good.

Almost 20 years ago.. damn.

https://creativecommons.org/2006/02/14/yahoodesignpatternlib...

https://web.archive.org/web/20060221111812/http://developer....

They had a great comparison of the different behaviors leaderboards could encourage in users.


Not quite the same thing, but there's this incredible (open source) project called The Component Gallery that is basically just a repository of UI components across 93 (currently) different design systems. It's an incredible resource if you're building a component from scratch and either want some design inspo or technical advice. Many of the design systems have thorough guidelines for a11y/ARIA best practices that I've learned a ton from

https://component.gallery/


Amen! The terms "accordion" and "carousel" were really codified by the pattern library. Establishing a common vernacular definitely accelerates things.

Oh, the second link is amazing. I love the old web, and that brought a lot of nostalgia.

YUI was ahead of its time as well

And it spawned ExtJS. Which could have been React, but they messed up. Literally they built a faster Facebook app 'Fastbook' in 2012.

Short history lesson:

https://medium.com/hackernoon/the-rise-and-fall-of-ext-js-c9...

In August 2006, a guy by the name of Jack Slocum (now CEO of Alta5) began to blog of his experiments with YUI. Over time, these experiments became more complex and Jack would start to bundle them into what would later be named YUI-ext (Yahoo UI extensions) — the precursor to Ext JS (Extensible JavaScript).

Jack Slocum’s blog was used to communicate his vision for YUI-ext and garnered community support from around the world. The release of the Grid component for YUI-ext would forever change the trajectory of the library and the community as the GridPanel would become the core UI component for many applications to come.

Throughout its early life, Jack continued to build upon YUI-ext by adding features to the framework, such as animations, Modals, Tab Panel, Resizable elements and a layout manager that greatly expanded upon the YUI framework. These components would seldom extend the YUI library and had their own rendering functions.

YUI-ext created a foundation for web programmers unlike anything the world had seen before and many developers flocked to the framework and invested in the newly formed community. The net result was the explosive expansion of YUI-ext.

From YUI-ext to Ext JS In January 2007 we found Jack extremely busy to push out YUI-ext 0.40 and it is this version where we find the namespace of the framework change from YAHOO.ext to a much simpler Ext (pronounced “ekst J S” or “E-X-T J S” by some of us old-school community members).

February 2007, Ext JS 1.0 was being developed in tandem with a new website, ExtJS.com. In April 2007, the launch of ExtJS.com was announced to the community along with the release of Ext JS 1.0.

https://web.archive.org/web/20230222210535/https://jackslocu...

For those that don’t know, Ext JS was one of the first JavaScript frameworks in the early days of Web 2.0. It was the first framework to offer a complete package of everything needed to build full-fledged applications using just JavaScript in a web browser. At one point, it was used by over 2 million developers worldwide, 70% of the fortune 500, and 8 of the top 10 financial institutions. It was years ahead of everyone else, open source, and had an incredible community of passionate people contributing to its success.

As that success grew, so did the number of copycat competitors. They eventually started taking the code and assets and embedding them into their own frameworks. Adobe embedded it in Cold Fusion and other competitive frameworks followed their lead without any contribution to the framework or community.

At the time the thought of competing directly against a behemoth like Adobe was scary. How could they take our product and offer it as their own? I took what I thought was the right action to “protect” Ext JS from being “stolen” by changing to a more restrictive license. That was a huge mistake.

Looking back in hindsight, without the fear, I have a much clearer picture. I see what truly made Ext JS great was not the code - it was all the people who loved, contributed and supported it. As we worked on making our own dreams a reality, we helped others do the same, sharing our knowledge, code, and solving tough challenges together.

That is what really mattered — our community. That is what I should have protected, not the code. You were my closest friends. I am sorry I changed the license after we all came to an agreement on the first license. That was a breach of integrity and you deserved better. I would do it differently if I could.


we often forget how great Yahoo engineering was back in the day, sad it was destroyed by bad management and horrible business cases prioritization

for some reason I remember him being related to YUI, but I learned JS from Douglas Crockford, one of the best lectures from the old days of JS.

You literally cannot buy Antigravity with a non-personal Google account.

I read someone on here who is using Gemini via OpenRouter because it was the only way they could pay for it.


I'm using it fine through both aistudio and vertex ai, direct API calls

It's not at all hard generally, the core of this issue is centered around gemini-cli which is a hot pile of trash. The inability to get keys or account credentials (like why even use an API key, Google is top notch in auto-auth/WIF)

Insanity to me how gemini-cli is so bad at the basics with so many great Google packages in open source that handle all this transparently. All I need to do is have my gcloud authd with the right account/project. I sarcastically assume his is because they vibe coded gemini-cli and it implemented everything from scratch, missing out on reusing those great packages


> I'm using it

Do you mean Antigravity or Gemini?

If you mean Antigravity then.. how? Their docs say you can't do this.

If you mean Gemini then I personally haven't had issues but haven't tried to productionize a Gemini app. The OPs account seems to reflect other comments here.


I already said how I'm calling Gemini

> direct API Calls

I suspect Antigravity to be a big flop like gemini-cli. They are so bad in this area they couldn't even write an extension or fork oss-code, instead spending $2B to pork an open source project with someone else's branding


Worth pointing out that South Korea had very limited democracy until the late 1980s: https://en.wikipedia.org/wiki/History_of_South_Korea

It's different almost by definition?

Because it was a once (twice!) off the impact and significance of it is amplified.


Wait - you think the solution to some people having a lower standard of living and others being persecuted is to kill them all?

Nukes usually don't wipe out entire countries, especially tactical nukes.

I'm far from convinced that using nukes in the Korean War would've been a good move, but equating it with "kill[ing] them all" is completely dishonest. What's your goal in this debate, and is it served by dishonest rhetoric?


My comment is in the context of:

> That USA didn't nuke China in 1950 or 1951. Would have solved a lot of problems for generations of people.

> USA dropping nukes probably would have been the better outcome for humanity.

Both of which I read as an expansive campaign of "nuking China"


In what way?

The US nearly lost the Korean war.

The US army was nearly overrun at least once.

The US airforce never achieved air superiority, and Soviet aircraft were better in most ways.

The only undisputed advantage the US had was nukes, which is why MacArthur wanted to use them tactically (!)


The subsequent Vietnam war reinforced this even more.

The only path that America had to win in Vietnam was to destroy it, including the population they were allegedly there to protect. Hence they lost.


This is simply not true.

Modern LLMs are trained by reinforcement learning where they try to solve a coding problem and receive a reward if it succeeds.

Data Processing Inequalities (from your link) aren't relevant: the model is learning from the reinforcement signal, not from human-written code.


Ok, then we can leave the training data out of the input, everybody happy.

I mean you'd think so, but...

> In fact, the UL2 20B model (at Google) was trained by leaving the job running accidentally for a month.

https://www.yitay.net/blog/training-great-llms-entirely-from...


Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: