Dot-Net, or Not Yet? (转)

Dot-Net, or Not Yet? (转)[@more@] Dot.NET, or Not Yet?

Yesterday, I was talking with one of VCDJ's frequent contributors, brainstoRming about the next article he ought to write. "How about something on C#?" I asked.

"Uhh," he replied, "the fact is I haven't written a single line of C# code yet."

"OK, no sweat," I said. "Want to do a piece on Managed Code?"

"Haven't gotten to that yet."

"Fine. Interested in writing about using C++ within the .net framework?"

"Look," he finally blurted. "I'm not even thinking about .NET yet. I've got to concern myself with writing code for products that have got to ship now. Alpha and betaware are just going to have to wait."

You've probably got a lot in common with this author. That is, like him, you've got a lot of demands on your time, and almost all of them have to do with getting your product out the door. And while your employer (whether that employer is a company or yourself) realizes you've got to stay on top of new developments, that gets ?shall we say, deprioritized in the face of looming code-complete dates.

This learn-what's-coming-up-but-use-what-you've-got dilemma is hardly new to developers. It's pretty well accepted, however, that VS.NET will cause the biggest waves in the development community since ?since ?sheesh, since ever. You can't wait 'til it ships to learn about it—there's too much there. So whatcha gonna do? Here's my advice:

Talk to your boss. Grab the last few issues of this magazine and march into your boss's Office. Spread the covers on the desk, point out that they all have something to do with .NET, and explain that .NET isn't just about the way you write code, it's about fundamental changes in the way applications work, as well as how they work together. Now—before it's time to start writing code to this new paradigm—is the time to think about how your applications fit into this framework.

Make it clear that both your job satisfaction and the company's relevance are on the line. Get a budget for books. Get time allocated for studying. Get the company to send you to conferences (check out http://www.vcdc.com/). Essentially, do whatever it takes to get some time and money set aside for learning about .NET. You and the company are going to have to climb the .NET curve. Start now, or you won't have momentum when it matters.

Get the preview. The technology preview (available now) is a free DOWNLOAD at microsoft.com/downloads/">http://msdn.microsoft.com/downloads/. Get comfortable with it.

Read this magazine. I've already mentioned that the last few covers of VCDJ have emphasized .NET. That's not going to stop any time soon. And while VCDJ shouldn't be your only source of .NET knowledge, it's a mighty fine place to start. Our .NET coverage is comprehensive, without being overwhelming. Don't worry, though, we are not—repeat, not—abandoning VS6. We'll continue covering VC++ 6 as long as you keep using it.

Experiment. True to VCDJ heritage, our .NET articles focus on how-to. Starting in this issue—and continuing through January—we're walking you through creating an e-commerce site using C# and other .NET technologies. Follow the steps, then play around with the project, adding features of your own. Let me know what you come up with: e-Mail vcdjedit@fawcette.com.

Decide what matters to you. You know your business best. There may well be parts of VS.NET or Microsoft's .NET strategy you can safely ignore. Consider how—and whether—these new technologies might affect you and your company, then focus on learning what you need to. You'll get to the other stuff when you have time. One additional thought along this line, though: Just because you don't need to know C# in your current job doesn't mean you won't need to know it in your next. Stay up-to-date; you'll be glad you did.

Don't freak out. It'll be months before VS.NET ships; you've still got time to get your arms around it. Learn a little at a time. Before long, as you're writing your VC++ code for today's projects, you'll see how you'd do the same thing—but better—in VS.NET. And once again, you're ahead of the curve. Nice going.


来自 “ ITPUB博客 ” ,链接:http://blog.itpub.net/10752043/viewspace-987687/,如需转载,请注明出处,否则将追究法律责任。

转载于:http://blog.itpub.net/10752043/viewspace-987687/

树莓派语音播报IP地址:Shell脚本与systemd服务实战 在嵌入式开发与Linux系统管理中,网络配置与调试是基础且关键的环节。其核心原理在于操作系统通过内核网络栈管理网络接口,并借助DHCP等协议动态获取IP地址。掌握网络信息查询命令(如`ip`与`ifconfig`)和文本语音(TTS)技术,能极大提升设备部署与远程管理的效率,尤其适用于无头(Headless)服务器、物联网设备快速调试等场景。本文以树莓派为硬件平台,深入讲解如何通过编写健壮的Shell脚本,结合`espeak`等TTS工具,实现开机自动语音播报IP地址,并封装为可靠的systemd服务,为 阅读详情

相关推荐

雅思听力词组总结

雅思听力词组总结(涵盖几乎所有雅思真题词组)已更新至1037条

weixin_69884785的博客 2463

Tool for Flatten A Folder

命令行工具。拷贝多个目录下的指定文件到目标目录

Distilling the Knowledge in a Neural Network 神经网络中的知识蒸馏

提高几乎任何机器学习算法性能的一种非常简单的方法是,在相同数据上训练许多不同的模型,然后对它们的预测结果取平均值[3]。遗憾的是,使用整个模型集成进行预测较为繁琐,并目计算成本可能过高,难以部署给大量用户,尤其是在各个模型都是大型神经网络时。Caruana及其合作者[1]已表明,可以将一个集成中的知识压缩到单个模型中,从而使其更易于部署;我们使用一种不同的压缩技术进一步发展了这种方法。

u013250861的博客 40

VSCode 新Unity调试插件(2023.8.8)问题

第一行应该是: 如果是还不行:只保留 其他内容删除,然后关闭vscode,unity内重新打开。一直Downloading the .NET Runtime.修改设置 setting.json 添加如下。没有这个配置可以通过【运行和调试】 进行创建。type: vstuc:就是unity。

qq_34191439的博客 4165

Dot-Net, or Not Yet?

Dot-Net, or Not Yet? Yesterday, I was talking with one of VCDJs frequent contributors, brainstorming about the next article he ought to write. "How about something on C#?" I asked. "Uhh," he repl

xu_yun的专栏 1078

Given a “flatten” dictionary object, whose keys are dot-separated

最近面试做了这样一个题:'''Given a “flatten” dictionary object, whose keys are dot-separated. For example, { ‘A’: 1, ‘B.A’: 2, ‘B.B’: 3, ‘CC.D.E’: 4, ‘CC.D.F’: 5}. Implement a function in any language to transfo...

guozehui123的博客 3160

Unity打包报错the given key was not present in the dictionary

在mac下打包项目时报错: error building player:KeyNotFoundException:the given key was not present in the dictionary 用Relpace而不是append的方式生成xcode项目,问题解决...只是要重新配置xcode了。

鱼蛋的开发小站 1万+

Dictionary Object

Dictionary Object VBScript的Dictionary 对象与PERL的关联数组(associative array)类似,其中存储的内容可以是任何数据,每一个存储的数据都与一个唯一的键(Key)关联,键用于获取每一项数据,键通常是整型或字符串类型。 'create the Dictionary object Set dict=CreateObject("Scripting

victoria_vicky的专栏 520

Distilling the Knowledge in a Neural Network

提高几乎任何机器学习算法性能的一种非常简单的方法是,在相同数据上训练许多不同的模型,然后对它们的预测结果取平均值[3]。遗憾的是,使用整个模型集成进行预测较为繁琐,并目计算成本可能过高,难以部署给大量用户,尤其是在各个模型都是大型神经网络时。Caruana及其合作者[1]已表明,可以将一个集成中的知识压缩到单个模型中,从而使其更易于部署;我们使用一种不同的压缩技术进一步发展了这种方法。

u013250861的博客 40

视觉特征-自监督-同源不同crop视图一致性【2021-05】:VICReg【同图➜双视图➜特征不变性】【批次每行一图、每列一特征维度】【正则化:列内方差防坍塌+列间协方差去冗余】【无负样本】

Abstract摘要最近的图像表征学习自监督方法通过由输入的编码器生成的之间的来进行学习。主要挑战是防止发生坍塌,即编码器生成恒定或无信息的向量。我们提出了与大多数解决相同问题的其他方法不同,并且在多项下游任务上取得了与最先进方法相当的结果。此外,我们表明,方差正则化项能够稳定其他方法的训练,并带来性能提升。

u013250861的博客 158

视觉特征-自监督-掩码特征预测【2024-11】:D-JEPA·T2I(通过下一词元预测实现高分辨率图像合成)【相对于D-JEPA的优化:文本条件、流匹配损失、VoPE】

近年来,自回归模型在类别条件图像生成方面展现了卓越的性能。然而,将下一词元预测应用于高分辨率文本到图像生成仍未得到充分探索。在本文中,我们提出了基于连续词元的自回归模型D-JEPA·T2I,该模型在架构和训练策略上均进行了创新,能够在高达4K的任意分辨率下生成高质量、照片级真实感的图像。在架构方面,我们采用去噪联合嵌入预测架构(D-JEPA),同时利用多模态视觉Transformer有效融合文本和视觉特征。此外,我们引入流匹配损失以及所提出的视觉旋位置编码(VoPE),以实现连续分辨率学习。

u013250861的博客 132

Fast R-CNN

本文提出了一种用于目标检测的 Fast Region-based Convolutional Network 方法(Fast R-CNN)。Fast R-CNN 在先前工作的基础上,利用深度卷积网络高效地对候选区域进行分类。与先前工作相比,Fast R-CNN采用了多项创新,在提高训练和测试速度的同时,也提升了检测准确率。Fast R-CNN 对非常深的 VGG16 网络的训练速度比 R-CNN 快 9×,测试时快 213×,并且在 PASCAL VOC 2012 上取得了更高的 mAP。

u013250861的博客 145

【202503】Wan:开放且先进的大规模视频生成模型【1.3B、14B】【支持对无限长度视频进行稳定的Inference】【Wan中的主要计算开销来自注意力机制】

wan: Open and Advanced Large-Scale Video Generative ModelsWan:开放且先进的大规模视频生成模型Wan Team, Alibaba Group Wan 团队,Alibaba GroupThis report presents Wan, a comprehensive and open suite of video foundation models designed to push the boundaries of video generation

u013250861的博客 585

PointWorld: 动作条件的全场景3D点flow 预测(用同一种3D物理空间模态来表示状态和动作)

PointWorld: Scaling 3D World Models for In-The-Wild Robotic Manipulation

u013250861的博客 230

High-Resolution Image Synthesis with Latent Diffusion Models 基于潜在扩散模型的高分辨率图像合成

Robin Rombach 1 * Andreas Blattmann1 * Dominik Lorenz 1 Patrick Esser ® Björn Ommer 1 1 Ludwig Maximilian University of Munich & IWR, Heidelberg University, Germany ° Runway ML慕尼黑路德维希-马克西米利安大学 & 德国海德堡大学 IWR ° Runway MLhttps://github.com/CompVis/latent-diff

u013250861的博客 82

【十万字拆解】从零开始实现GPT-2(完整版)

本文是一篇 14 万字、从零硬核实现 GPT 的保姆级指南。内容全面覆盖四个阶段:①从最基础的 Bigram 语言模型切入,跑通数据处理与训练闭环;②仿照 Transformer 架构,深度研究 Multi-Head Attention、残差网络、LayerNorm 及 Dropout 的底层数学与代码实现;③从零构建BPE 分词器,对齐 GPT-2 规范;④从头搭建 GPT-1/2 工业级预训练流程,包含混合精度、梯度累积等十余项硬件级工程调优,并在 AutoDL 算力平台上从0开始训练!适合所有层次!

Pi_Nick的博客 1029

High-Resolution Image Synthesis with Latent Diffusion Models # 基于潜在扩散模型的高分辨率图像合成

Robin Rombach 1 * Andreas Blattmann1 * Dominik Lorenz 1 Patrick Esser ® Björn Ommer 1 1 Ludwig Maximilian University of Munich & IWR, Heidelberg University, Germany ° Runway ML慕尼黑路德维希-马克西米利安大学 & 德国海德堡大学 IWR ° Runway MLhttps://github.com/CompVis/latent-diff

u013250861的博客 56

CV-骨干网络01:Swin Transformer(使用移位窗口的分层视觉Transformer)【通过patchmerging在不同stage逐步降低分辨率、提高通道数,输出一组“多尺度特征图”】

2.7+ 2 . 72.7box AP and2.6+ 2 . 62.63.2+ 3 . 23.2本文提出了一种名为Swin Transformer的新视觉Transformer,它能够作为计算机视觉的通用骨干网络。将Transformer从语言领域迁移到视觉领域所面临的挑战源于这两个领域的差异,例如视觉实体尺度的巨大变化以及图像中像素相对于文本中单词的高分辨率。为了解决这些差异,我们提出了一种分层Transformer,其表征通过Shiftedwindows(移位窗口)进行计算。

u013250861的博客 59
上一篇: 充分利用Windows API扩展Delphi函数 (转)
下一篇: Win32 API资源分配释放速查,防止代码资源泄露 (转)
csd3176
博客等级 码龄11年 20粉丝 0原创
评论
成就一亿技术人!
拼手气红包6.0元
还能输入1000个字符  | 博主筛选后可见
 
 条评论被折叠 查看
添加红包

请填写红包祝福语或标题

红包个数最小为10个

红包金额最低5元

当前余额3.43前往充值 >
需支付:10.00
成就一亿技术人!
领取后你会自动成为博主和红包主的粉丝 规则
hope_wisdom
发出的红包
实付
使用余额支付
点击重新获取
扫码支付
钱包余额 0

抵扣说明:

1.余额是钱包充值的虚拟货币,按照1:1的比例进行支付金额的抵扣。
2.余额无法直接购买下载,可以购买VIP、付费专栏及课程。

余额充值